letters
npg
© 2014 Nature America, Inc. All rights reserved.
An epigenetic mechanism of resistance to targeted therapy in T cell acute lymphoblastic leukemia Birgit Knoechel1–4,16, Justine E Roderick5,16, Kaylyn E Williamson1,2,6,7, Jiang Zhu1,2,6,7, Jens G Lohr2,8, Matthew J Cotton1,2,6,7, Shawn M Gillespie1,2,6,7, Daniel Fernandez2,9, Manching Ku1,2,6, Hongfang Wang10, Federica Piccioni2, Serena J Silver2, Mohit Jain2,11,12, Daniel Pearson4,13, Michael J Kluk10, Christopher J Ott8, Leonard D Shultz14, Michael A Brehm15, Dale L Greiner15, Alejandro Gutierrez3,4, Kimberly Stegmaier3,4, Andrew L Kung3,4, David E Root2, James E Bradner2,8, Jon C Aster10, Michelle A Kelliher5 & Bradley E Bernstein1,2,6,7 The identification of activating NOTCH1 mutations in T cell acute lymphoblastic leukemia (T-ALL) led to clinical testing of g-secretase inhibitors (GSIs) that prevent NOTCH1 activation1–4. However, responses to these inhibitors have been transient5, suggesting that resistance limits their clinical efficacy. Here we modeled T-ALL resistance, identifying GSI-tolerant ‘persister’ cells that expand in the absence of NOTCH1 signaling. Rare persisters are already present in naive T-ALL populations, and the reversibility of their phenotype suggests an epigenetic mechanism. Relative to GSI-sensitive cells, persister cells activate distinct signaling and transcriptional programs and exhibit chromatin compaction. A knockdown screen identified chromatin regulators essential for persister viability, including BRD4. BRD4 binds enhancers near critical T-ALL genes, including MYC and BCL2. The BRD4 inhibitor JQ1 downregulates expression of these targets and induces growth arrest and apoptosis in persister cells, at doses well tolerated by GSI-sensitive cells. Consistently, the GSI-JQ1 combination was found to be effective against primary human leukemias in vivo. Our findings establish a role for epigenetic heterogeneity in leukemia resistance that may be addressed by incorporating epigenetic modulators in combination therapy. T-ALL is an aggressive malignancy frequently associated with activating mutations in NOTCH1, a critical oncogene in this disease1. GSIs that prevent NOTCH1 cleavage and activation have been tested in clinical trials and mouse models, but responses have been modest and
transient5. To understand the mechanisms by which T-ALL cells overcome NOTCH1 inhibition, we modeled GSI resistance in vitro and in vivo and investigated the functional and molecular characteristics of resistant cells. By chronically exposing NOTCH1-dependent human T-ALL cells to a GSI (compound E) in vitro, we isolated a population of persister cells that tolerated GSI concentrations more than 50-fold higher than the concentrations tolerated by naive cells (Fig. 1a,b and Supplementary Figs. 1a and 2a). To characterize these drug-tolerant cells, we first examined the active intracellular form of NOTCH1 (ICN1)6. ICN1 was present at high levels in naive T-ALL cells but was essentially undetectable in persister cells (Fig. 1c, top and Supplementary Fig. 2b, top). Expression of NOTCH1 target genes, including DTX1 and HES4, was also downregulated in persister cells (Supplementary Figs. 1b,c and 3i). The phenotype was reversible, as persister cells rapidly re-expressed ICN1 and NOTCH1 target genes after removal of the GSI (Fig. 1c and Supplementary Figs. 1b,d and 2b). Re-exposure of these ‘reversed’ cells to GSI led to downregulation of the expression of NOTCH1 target genes and growth arrest, as in the original naive T-ALL population (Supplementary Fig. 1b). This reversibility suggests that the resistance phenotype is mediated epigenetically. Induction of MYC is thought to be a major mechanism whereby constitutive NOTCH1 activation results in leukemic T cell transformation7–9. Although MYC protein levels were dramatically reduced by short-term GSI treatment, moderate levels were maintained in the persister cells, despite continued GSI treatment (Fig. 1c and Supplementary Fig. 2b). Consistently, the strong MYC transcriptional
1Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA. 2Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA. 3Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA. 4Division of Hematology/Oncology, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, USA. 5Department of Cancer Biology, University of Massachusetts Medical School, Worcester, Massachusetts, USA. 6Center for Cancer Research, Massachusetts General Hospital, Boston, Massachusetts, USA. 7Howard Hughes Medical Institute, Chevy Chase, Maryland, USA. 8Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA. 9Biostatistics Graduate Program, Harvard University, Cambridge, Massachusetts, USA. 10Department of Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, USA. 11Center for Human Genetic Research, Massachusetts General Hospital, Boston, Massachusetts, USA. 12Department of Molecular Biology, Massachusetts General Hospital, Boston, Massachusetts, USA. 13Biological and Biomedical Sciences Graduate Program, Harvard Medical School, Boston, Massachusetts, USA. 14Jackson Laboratory, Bar Harbor, Maine, USA. 15Program in Molecular Medicine, University of Massachusetts Medical School, Worcester, Massachusetts, USA. 16These authors contributed equally to this work. Correspondence should be addressed to B.E.B. ([email protected]) or M.A.K. ([email protected]).
Received 12 August 2013; accepted 6 February 2014; published online 2 March 2014; doi:10.1038/ng.2913
364
VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
letters
1 0 Naive
Persister
d
0.4
mTOR
MYC
0.2
Tubulin
0
Tubulin
P
Naive Persister
1.2 1.0 0.8 0.6 0.4 0.2 0
N ST
0
0.625 1.25 2.5 5 Rapamycin (nM)
10
Figure 1 A subpopulation of drug-tolerant T-ALL cells mediates Persister Naive resistance to NOTCH1 inhibition. 1.0 Reversed 1.0 Rapamycin pretreated 1.0 Naive (a) Bar plot comparing the expansion 0 0 0.8 0.8 0.8 rates of native T-ALL cells (grown in the absence 0.6 MYC 0.6 0.6 of GSI) and persister cells grown 0.4 0.4 0.4 continuously in 1 µM GSI (two Tubulin replicates; error bars, s.d.). 0.2 0.2 0.2 *** *** (b) Proliferation of naive and *** 0 0 0 persister cells treated with Control Treated Control Treated Control Treated the indicated doses of NOTCH inhibitor for 6 d (two replicates; error bars, s.d.). (c) Protein blot showing ICN1 and MYC levels in naive cells (N), cells with short-term treatment (ST; 5 d of 1 µM GSI), persister cells in 1 µM GSI (P), reversed persister cells removed from GSI for 2 weeks (Rev) and reversed cells re-exposed to GSI (Rev tx; 5 d). Reactivation of NOTCH1 signaling in reversed cells suggests that the persister phenotype is reversible. (d) Protein blot showing phosphorylated mTOR (p2481), a marker of activated mTOR signaling, total mTOR and tubulin in naive (N) and persister (P) cells. (e) Proliferation of naive and persister cells treated with the indicated concentrations of rapamycin for 6 d (four replicates; error bars, s.d.). (f) Protein blots showing MYC expression in naive and persister cells after 3 d of treatment with 2 µM AKT inhibitor MK-2206 (AKTi) or 10 nM mTOR inhibitor rapamycin (Rapa). (Data shown are for DND-41 cells; data for KOPT-K1 cells are shown in Supplementary Fig. 2). (g) Rare drug-tolerant cells pre-exist in naive T-ALL populations. Single cells from a naive DND-41 population, a reversed population (reversed for 1 month) and a naive population pretreated with 10 nM rapamycin for 3 d were sorted into individual wells of 96-well plates and cultured in 1 µM GSI (treated) or without drug (control) for 4–6 weeks. Bar plots show the fraction of single cells that form colonies (sorted single cells per condition: naïve, n = 936; reversed, n = 936; rapamycin pretreated, n = 858; pooled from 2 independent experiments; ***P < 0.001; error bars, s.d.). These data suggest that GSI tolerance is already evident in a small fraction of naive T-ALL cells and is reversible.
signature of naive T-ALL cells was reduced in persister cells (Supplementary Table 1). Conversely, gene signatures associated with mitogen-activated protein kinase (MAPK), c-Jun N-terminal kinase (JNK), phosphoinositide 3-kinase (PI3K) and mTOR signaling were enhanced (Supplementary Table 1). Genetic alterations of PTEN and consequent AKT pathway activation have been associated with GSI resistance10,11. Although the T-ALL cells modeled here did not harbor PTEN mutations11, persister cells had higher levels of phosphorylated PTEN and were sensitive to AKT inhibitors (Supplementary Figs. 1e and 2d). mTOR activation has also been observed in T-ALL 12,13. Persister cells had higher levels of p2481 (phosphorylated mTOR) and were sensitive to the mTOR inhibitor rapamycin (Fig. 1d,e and Supplementary Fig. 2c). Inhibition of mTOR markedly reduced MYC protein levels in persister cells but not in naive T-ALL cells (Fig. 1f). In contrast, persister cells did not exhibit enhanced sensitivity to doxo rubicin or vorinostat (Supplementary Figs. 1f and 2e). These data suggest that rewired signaling programs maintain MYC expression and cell proliferation in the absence of NOTCH1 activity. Next, we sought to determine whether resistance is induced by GSI treatment or whether, alternatively, a fraction of unexposed cells is already drug tolerant. In support of the latter hypothesis, when we isolated single cells from a naive T-ALL population, we found that ~4% expanded in the presence of GSI (Fig. 1g). In contrast, the remaining cells failed to grow, even after extended periods of time. An analogous analysis of reversed cells, which were exposed to GSI but then expanded in the absence of drug for 1 month, again showed a small fraction of GSI-tolerant single cells (~9%). We hypothesized that pre-existing GSI-tolerant cells share phenotypic characteristics with persister cells, such as mTOR dependency. In support of this Nature Genetics VOLUME 46 | NUMBER 4 | APRIL 2014
Fraction of single-cell clones
Fraction of single-cell clones
Fraction of single-cell clones
AK
AK
Ti R ap a
g
Ti R ap a
f
© 2014 Nature America, Inc. All rights reserved.
p2481
P R e R v ev tx
0.6
GSI (µM)
npg
N
Tubulin
0.8
e
ICN1
P R e R v ev tx
2
1.0
c
N ST
3
1.2
50 10 0
Cell survival normalized to control (CellTiter-Glo)
Fold expansion in 3 d
4
Naive Persister
Cell survival normalized to control (CellTiter-Glo)
b
0 0. 4 0. 8 1. 5 3. 1 6. 2 12 .5 25
a
idea, pretreatment with rapamycin reduced the frequency of GSItolerant single cells by ~5-fold (Fig. 1g). These data suggest that GSI resistance depends on a pre-existing subpopulation of T-ALL cells with distinct signaling programs that readily expand in the absence of NOTCH1 signaling. The reversibility of the phenotype suggests that this heterogeneity is mediated by epigenetic rather than genetic alterations. In addition to altered signaling, persister cells exhibited morphologic changes, including decreased cell and nuclear sizes (Fig. 2a,b and Supplementary Fig. 3a,b). We postulated that these changes might reflect global chromatin changes associated with NOTCH1 inactivity. We found that the persister cells had higher global levels of repressive histone modifications and heterochromatin protein 1 (HP1) (Fig. 2c,d and Supplementary Fig. 3c,e). Chromatin in persister cells was also relatively inaccessible to micrococcal nuclease (MNase) digestion (Supplementary Fig. 3g), consistent with a compact state. Chromatin compaction likely reflects the absence of NOTCH1 activity, as it was also evident in cells with short-term GSI treatment and in T-ALL cells in which NOTCH1 signaling was inhibited by dominant-negative mastermind-like 1 (DN-MAML1) (Fig. 2c and Supplementary Figs. 1c and 3c,e,i). We postulated that the altered chromatin state in persister cells might uncover new susceptibilities that could be targeted by emerging epigenetic therapies. To test this hypothesis, we designed a lentiviral short hairpin RNA (shRNA) knockdown screen that targeted ~350 chromatin regulators with an average of 5 independent hairpins per gene (Online Methods). We identified ~15 genes for which knockdown compromised the survival of naive and persister cells, which were thus presumed to be generally required in T-ALL 365
npg
H3K9me3/total H3
H3K27me3/total H3
600,000 400,000 200,000
0.050
** 0.025
0
N ST P BRD4 Tubulin
1.0
0.20
tx
1
0
0.25
ut
0.5
JQ
0.0625 0.125 0.25 JQ1 (µM)
o
0
N
0
ho
SND1 SS18 STAG3L1 SUDS3
0.01
800,000
0.075
as
PRMT2 EZH2 SUV39H1 PRMT5 PRMT7 KDM5D MLL5 MINA MORF4L1
0.02
1,000,000
w
te
ai
is rs
Pe
N
r
0
ARID3B BRD1 BRD4 CD2BP2 CHD1 CHD8 CHRAC1
0.03
1,200,000
0.100
1
0.05
**
0
e-
Preferential dependency persister
0.04
5
0
pr
0.10
0.05
0.5
h
1,400,000
10
1
0.15
2.5×
15
JQ
0.20
0.06
1.0
**
20
JQ
0.25
HMGXB4 PRDM4 HMG20B PRMT3 MBD6 PRMT8 SMOX SPTY2D1
***
1.5
Naive Persister
1,600,000
Sh N tre or aive at t-te m r Pe en m rs t is te r
**
HDAC2 HDAC4 HDAC5 HDAC6 HDAC9
BRD4 mRNA normalized to B2M
0.30
ve
CBX3 mRNA normalized to B2M
© 2014 Nature America, Inc. All rights reserved.
Preferential dependency naive ARID5A BAZ1B CHAF1B CHMP4A CHD5 CHD7 EHMT2
te r
g 0.07
2.0
Sh Na i tre ort ve at -te m rm e Pe nt rs is te r
f
*
25
Fraction of single cell clones normalized to untreated
e
500 300
30
**
N Sh aiv tre ort e at -te m rm e Pe nt rs is te r
d
600
is
0 200 400 600 800 1,000 FSC
700
e
0
800
ai v
20
350
*
2.5
Pe rs
40
400
900
N
60
Nuclear area (pixels)
80
Persister
Cell survival (CellTiter-Glo)
Naive Persister
c 3.0 Naive
Fraction of apoptotic cells (Caspase-Glo/CellTiter-Glo)
Percentage of maximum
100
b
450
N ai Pe ve rs is te R r ev er se R d et re at ed
a
FSC (median fluorescence intensity)
letters
0.15
Figure 2 Drug-tolerant T-ALL cells adopt an altered 0.10 chromatin state and are BRD4 dependent. (a) Left, forward scatter (FSC) distribution of naive and persister cells determined by flow cytometry; right, bar 0.05 plot of FSC indicating the size distributions of naive (blue), persister (red) and 0 reversed (green) cells and reversed cells re-exposed to GSI for 5 d (retreated; orange). 0 0.0625 0.125 0.25 0.5 1.0 (b) Sizes of naive and persister cell nuclei are shown by 4,6-diamidino-2-phenylindole JQ1 (µM) (DAPI) staining (left; scale bars, 50 µm) and quantified in the box plots (right; naive, n = 585; persister, n = 496; P < 1 × 10−4; boxes indicate 25th, 50th and 75th percentiles, whiskers represent 5th and 95th percentiles). (c) Bar plots indicating relative levels of repressive histone modifications per ELISA of bulk histones from naive cells undergoing short-term treatment (3 d, 1 µM GSI) and persister cells (two (technical) replicates; error bars, s.d.; *P < 0.05, **P < 0.01). (d) CBX3 mRNA (encoding HP1γ) expression shown for naive and persister T-ALL cells (two replicates; error bars, s.d.; **P < 0.01). (e) The shRNA screen identified chromatin regulators preferentially required for naive or persister cell survival. Top hits for each cell state are indicated. (f) BRD4 expression is shown for naive cells (N), cells undergoing short-term treatment (ST; 5 d) and persister cells (P) (top, mRNA expression; bottom, protein expression; three (technical) replicates; error bars, s.d.; **P < 0.01, ***P < 0.001). (Data shown are for DND-41 cells; data for KOPT-K1 cells are shown in Supplementary Fig. 3). (g) Proliferation (6 d of treatment; four replicates) and rate of apoptosis (4 d of treatment; three replicates) are shown for naive and persister cells treated with the indicated doses of JQ1 (error bars, s.d.). (h) Naive DND-41 populations were pretreated with 0.5 µM JQ1 for 4 d (JQ1 pre-tx), pretreated with 0.5 µM JQ1 for 4 d but then cultured for 24 h without JQ1 (JQ1 washout) or never exposed to JQ1 (no JQ1). Single cells from each of these populations were sorted into individual wells of 96-well plates and cultured with GSI (1 µM) or without GSI (control) for 6 weeks. The bar plot indicates the fraction of single cells that form colonies in GSI relative to control cells (sorted single cells per condition: n = 936; data pooled from 2 independent experiments; error bars, s.d. ; **P < 0.01). Pretreatment with JQ1 eliminates pre-existing GSI-tolerant cells from naive T-ALL populations, but this effect is reversed by 24 h of JQ1 washout.
(Supplementary Table 2). We also identified genes preferentially required for the survival of either naive or persister cells. For example, naive cells were preferentially dependent on several histone deacetylases, whereas persister cells were more dependent on certain arginine methyltransferases and histone methyltransferases (Fig. 2e and Supplementary Table 2). We identified BRD4 as a top hit in the screen, with three shRNAs noticeably reducing the proliferation of persister cells without affecting the proliferation of the naive population (Supplementary Fig. 4a,b and Supplementary Table 2). BRD4 is a member of the BET family of bromodomain proteins that bind acetylated histones14,15. BRD4 has been implicated in several malignancies, including acute myeloid leukemia and lymphoma16,17. BRD4 expression (mRNA and protein) increased with NOTCH1 inhibition and was higher in persister cells (Fig. 2f and Supplementary Fig. 3f). 366
This BRD4 dependency is of particular interest given the development of small molecule BET inhibitors, such as JQ1 (refs. 18,19). Indeed, we found that persister cells exhibited ~5-fold enhanced sensitivity to JQ1 and underwent proliferation arrest and apoptosis in response to JQ1 concentrations that were well tolerated by naive T-ALL cells (Fig. 2g and Supplementary Figs. 2e and 4c). We therefore considered whether the preexisting GSI-tolerant cells, identified in naive T-ALL populations by single-cell analysis (Fig. 1g), might also be BRD4 dependent. Indeed, when we pretreated naive T-ALL cells with JQ1 for 4 d and then removed the JQ1 immediately before isolating individual cells into 96-well plates, the frequency of GSI-tolerant single cells was reduced by ~20-fold (Fig. 2h). However, when we pretreated naive cells but then removed JQ1 24 h before isolating individual cells, we were able to detect GSI-tolerant VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
letters d
15 H3K36me3
80,000 60,000
RPTOR CLEC16A CDK6 ETV6 BCL2L1 BCL2
40,000 20,000 0 0
200
600
1,000
Genes ranked by BRD4 signal
BRD4 15 H3K36me3
e
Persister
H3K36me3 BRD4
hg19
hg19
10 kb 50 50 H3K36me3 BRD4 50
JQ1 (µM) 0 0.25 0.50
0 0.25 0.50
BCL2
Naive
Persister
H3K36me3 BRD4
200 kb
hg19
MYC
20 80
Tubulin
20
H3K36me3
Naive
3 2 1 Persister
g EV
50,000
BCL2 ORF
**
40,000 30,000
***
20,000
**
10,000
**
0
Persister
0
ETV6
h
0.0625 0.125 0.25 JQ1 (µM)
0.5
Figure 3 BRD4 binds enhancers near developmental, cell cycle and prosurvival genes in T-ALL. 100,000 EV (a) Heat map showing enrichment signals for BRD4, H3K27ac and H3K4me1 over 29,978 MYC ORF H3K4me1-marked distal sites in persister cells (rows represent 10-kb regions centered 80,000 on H3K4me1 peaks ranked by overall signal intensities for BRD4 and H3K27ac). (b) BRD4 peaks 60,000 were ranked by the signal intensity of the area under the curve (AUC) in persister cells, averaging BRD4 signal intensities of DND-41 and KOPT-K1 persister cells to identify candidate superenhancers. 40,000 The plot depicts top-ranked sites linked to an active gene in T-ALL (Online Methods). (c) Tracks * showing BRD4 binding and H3K36me3 enrichment (marking transcribed regions) over the CDK6 and 20,000 ETV6 loci, both of which contain BRD4-bound superenhancers. (d,e) Left, tracks showing BRD4 binding and H3K36me3 enrichment across the BCL2 (d) and MYC (e) loci in naive and persister cells. 0 Right, plots showing BRD4 enrichment as measured by ChIP and quantitative PCR (qPCR) over putative 0 0.0625 0.125 0.25 0.5 JQ1 (µM) regulatory elements in the BCL2 and MYC loci in naive and persister cells (two replicates; error bars, s.d.; BC, background control). (f) Protein blots showing BCL2 and MYC expression in naive and persister cells after 4 d of treatment with the indicated JQ1 doses. (Data shown are for KOPT-K1 cells; data for DND-41 cells are shown in Supplementary Fig. 5). (g) Plot showing relative proliferation of DND-41 persister cells transfected with BCL2 expression vector (BCL2 ORF) or empty vector control (EV) after 6 d of treatment with the indicated JQ1 doses (three replicates; error bars, s.d.; **P < 0.01, ***P < 0.001). (h) Plot showing relative proliferation of DND-41 persister cells transfected with MYC expression vector (MYC ORF) or empty vector control (EV) after 6 d of treatment with the indicated JQ1 doses (two replicates; error bars, s.d.; *P < 0.05). Cell survival (CellTiter-Glo)
© 2014 Nature America, Inc. All rights reserved.
npg
80 BRD4
4
Naive
30 CDK6
5
0
MYC
H3K36me3
5
6 MYC locus
H3K36me3
f
10
Naive Persister
BRD4
30 80
15
BCL2
Cell survival (CellTiter-Glo)
BRD4 Naive
200 kb
Persister
80
BCL2 locus
20
0 BCL2
50
c
25
BRD4 binding enrichment/BC
100,000
Persister
70
Naive
BRD4 signal (AUC)
H3K4me1 sites (n = 29,978)
hg19
200 kb 70 BRD4
BRD4 binding enrichment/BC
MSI2
120,000
Naive
b
BR D H 4 3K 27 a H 3K c 4m e1
a
clones in a proportion similar to that found in control naive T-ALL cells. Thus, pre-existing GSI-tolerant cells in naive T-ALL populations are highly BRD4 dependent, like persister cells, and exist in dynamic equilibrium with the more prevalent GSI-sensitive cells. To investigate its regulatory functions, we mapped BRD4 binding in naive and persister T-ALL populations by chromatin immunoprecipitation and sequencing (ChIP-seq). We also mapped histone modifications associated with promoters (trimethylation of histone H3 at lysine 4, H3K4me3), transcripts (trimethylation of histone H3 at lysine 36, H3K36me3), enhancers (monomethylation of histone H3 at lysine 4, H3K4me1 and acetylation of histone H3 at lysine 27, H3K27ac) and Polycomb-repressed loci (trimethylation of histone H3 at lysine 27, H3K27me3)20,21. In both cell states, BRD4 bound promoters (~30% of sites) and putative enhancers enriched for H3K4me1 and H3K27ac marks (~60%) (Fig. 3a,b and Supplementary Fig. 5a). We collated the largest BRD4 binding sites in persister cells, as Nature Genetics VOLUME 46 | NUMBER 4 | APRIL 2014
these might reflect ‘superenhancers’ with critical functions in cell type–specific gene regulation22. Proximal genes encoded many known T-ALL regulators, including the transcription factor ETV6, the cell cycle regulator CDK6, the signaling molecule RPTOR and the prosurvival protein BCL2 (Fig. 3b–d, Supplementary Fig. 5b,d and Supplementary Table 3). Although most large BRD4 peaks were shared by the populations, some were specific to either naive cells (for example, those in NOTCH1 target loci) or persister cells (Supplementary Fig. 5c). Thus, BRD4 binds and may sustain the activity of the regulatory elements and target genes required for the proliferation of T-ALL cells. We next considered the molecular basis for the preferential BRD4 dependency in persister cells. The relatively consistent BRD4 binding patterns in naive and persister cells suggest that BRD4 relocalization is unlikely to explain the different dependencies of these cells. However, persister cells exhibited an altered chromatin environment with greater 367
letters tions in potential regulatory regions (Supplementary Fig. 3h). BRD4 is believed to have an important role as a ‘bookmark’ of active regulatory elements, maintaining their chromatin state as cells progress through mitosis15,23. Thus, we suggest that generalized chromatin repression in
ICN1
Cell survival (CellTiter-Glo)
MYC
0.0015
0.0010
***
0.0005
0
V
100,000
0.020
0.015
0.010
0.005
0
LT
V
80,000 70,000 60,000
**
50,000 40,000
**
30,000
**
20,000 10,000 0
e NSG mice
25% leukemic blasts in the periphery
DBZ + JQ1
0.05 0.04 0.03 0.02 0.01 0
80,000
V
**
20,000 Veh GSI
100
4,000 3,000 2,000 1,000 0
V
LT
T-ALL-x-14
9,000
40,000
f Monitor for disease latency
**
*
5,000
LT
60,000
0
Administer treatment for 3 weeks
JQ1
0.06
*
100,000
GSI JQ1 GSI + JQ1
DBZ
0.07
T-ALL-x-11
Vehicle
1 × 106 cells
*
0.08
120,000
Percent survival
Primary leukemia
6,000
0.09
LT
140,000
T-ALL-x-9
90,000
0
8,000 7,000 6,000 5,000 4,000 3,000
**
2,000 1,000 0
JQ1 GSI + JQ1
0
GSI
JQ1 GSI + JQ1
T-ALL-x-9
80 60 40
368
Percent survival
20 Figure 4 Combination therapy targeting NOTCH1 and BRD4 in primary T-ALL. (a) NSG mice were 0 injected with luciferase-expressing KOPT-K1 T-ALL 0 20 40 60 80 100 120 140 160 cells, and leukemic mice were treated with Treatment NOTCH inhibitor DBZ for 5 d (ST; three doses), Days after transplantation DBZ for 3 weeks (LT; dosing every other day) or T-ALL-x-11 T-ALL-x-14 vehicle (V) (five mice per group). DBZ-treated 100 100 Vehicle mice were sacrificed when bioluminescence DBZ had plateaued and then significantly increased 80 80 JQ1 (Online Methods and Supplementary Fig. 7a). DBZ + JQ1 60 60 Images show hematoxylin and eosin (H&E) staining and immunohistochemistry for ICN1 40 40 and MYC in bone marrow from the respective leukemic mice. Scale bar, 100 µm. (b) NOTCH1 20 20 target gene expression shown for leukemia cells 0 0 sorted from the spleen of mice treated with vehicle (V) 0 20 40 60 80 100 130 150 170 190 0 10 110 or with long-term treatment (LT). Data points reflect Treatment Treatment averages for 2 mice (2 replicates; error bars, s.d.; Days after transplantation Days after transplantation ***P < 0.001). (c) Left, CBX3 expression shown for leukemia cells as in b (two replicates; error bars, s.d.; *P < 0.05). Right, intracellular BCL2 expression measured by flow cytometry shown for leukemia cells as in b (two replicates; error bars, s.d.; *P < 0.05). (d) Proliferation of three primary T-ALLs grown in vitro for 6 d in the presence of 1 µM GSI, 0.25 µM JQ1 or both compounds (T-ALL-x-9 and T-ALL-x-14: two replicates; T-ALL-x-11: three replicates; error bars, s.d.; *P < 0.05, **P < 0.01). (e) Experimental design for primary T-ALL xenotransplantation trials. Primary T-ALL samples from three primary pediatric T-ALLs (T-ALL-x-9, T-ALL-x-11 and T-ALL-x-14; Supplementary Table 4) were transplanted into NSG mice. Once leukemic burden reached 25% in peripheral blood by human CD45 staining, mice were randomized into four treatment groups (vehicle, JQ1, DBZ or DBZ-JQ1 combination). Mice were treated for three consecutive weeks and were then monitored for disease and sacrificed when they became moribund. (f) KaplanMeier survival curves for samples T-ALL-x-9, T-ALL-x-11 and T-ALL-x-14 (P < 0.01 as assessed by the log-rank test; for details see Supplementary Table 4; T-ALL-x-9: vehicle n = 5, DBZ n = 5, JQ1 n = 5, DBZ and JQ1 n = 4; T-ALL-x-11: vehicle n = 4, DBZ n = 5, JQ1 n = 5, DBZ and JQ1 n = 4; T-ALL-x-14: vehicle n = 5, DBZ n = 5, JQ1 n = 5, DBZ and JQ1 n = 4). Combination treatment significantly prolonged survival compared to single-agent treatment and treatment with vehicle in all three trials. Percent survival
npg
© 2014 Nature America, Inc. All rights reserved.
d
0.0020
c
0.025
BCL2 (mean fluorescence intensity)
H&E
0.0025
Cell survival (CellTiter-Glo)
b
LT
Cell survival (CellTiter-Glo)
ST
HES4 mRNA expression normalized to B2M
V
DTX1 mRNA expression normalized to B2M
a
CBX3 mRNA epxression normalized to B2M
compaction, higher levels of repressive histone modifications and lower levels of the enhancer-associated H3K27ac mark (Fig. 2c and Supplementary Fig. 3c–e,g). Consistently, ChIP-seq data showed that persister cells had modestly higher levels of repressive histone modifica-
VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
npg
© 2014 Nature America, Inc. All rights reserved.
letters persister cells renders enhancers particularly dependent on BRD4 for their epigenetic maintenance (Supplementary Fig. 6). We next focused on individual genes that might account for BRD4 dependency. BCL2 is an established BRD4-dependent gene in mixed-lineage leukemia 24. BCL2 was near a top-ranked BRD4 binding site and was expressed at higher levels in persister cells (Fig. 3b,d,f and Supplementary Figs. 3i and 5d). Treatment with JQ1 markedly reduced BCL2 protein expression in persister cells but had little effect on its expression in naive cells (Fig. 3f). Downregulation of BCL2 expression appears to contribute to the reduced survival of JQ1-treated persister cells, as exogenous BCL2 overexpression partially rescued these cells from JQ1 treatment (Figs. 2g and 3g). The MYC oncogene is an established driver in T-ALL and represents a canonical BRD4 target18. A large enhancer in the MYC locus was strongly enriched for BRD4 binding (Fig. 3e and Supplementary Fig. 5f). Treatment with JQ1 noticeably reduced MYC protein expression in persister cells at doses that did not alter MYC expression in naive cells (Fig. 3f). This effect could partially be rescued by MYC overexpression (Fig. 3h). Hence, compromised survival of JQ1-treated persister cells depends on the downregulation of BCL2 and MYC. To investigate the in vivo relevance of GSI resistance and associated epigenetic changes, we injected luciferase-expressing KOPT-K1 TALL cells orthotopically into NOD-scid IL2Rg null (NSG) mice and followed bioluminescence over time. GSI resistance developed rapidly in vivo after a short period of slowed leukemic growth (Supplementary Fig. 7a). ICN1 levels were drastically reduced in the bone marrow of GSI-treated mice, and expression of NOTCH1 target genes was downregulated in the leukemia cells, indicating that resistance is not due to NOTCH1 reactivation (Fig. 4a,b). These in vivo persister cells also shared other phenotypic characteristics with their in vitro counterparts, including reactivation of MYC expression and increased levels of CBX3 mRNA (encoding HP1γ) and BCL2 (Fig. 4a,c). We hypothesized that combined inhibition of NOTCH1 and BRD4 might be an effective treatment for T-ALL, including relapsed or induction failure disease, where conventional therapy has failed. We therefore examined the efficacy of GSI-JQ1 combination therapy in primary T-ALLs in vitro and in vivo. We examined three tumors from pediatric patients collected at diagnosis or upon relapse (Supplementary Fig. 7b and Supplementary Table 4), including a sample isolated from an individual with relapsed early thymic progenitor (ETP) T-ALL, who failed to respond to cytotoxic therapy. First, we examined the sensitivity of these leukemias to GSI, JQ1 or combined treatment in vitro. We found that the combination treatment was significantly more effective at inhibiting cell growth and survival than single-agent treatment (Fig. 4d). To examine effects in vivo, we engrafted NSG mice with the primary T-ALLs and initiated treatment when a leukemic burden of ~25% in peripheral blood was reached. Randomized groups were treated with vehicle, NOTCH inhibitor (DBZ), BRD4 inhibitor (JQ1) or the DBZ-JQ1 combination for 3 weeks, and the effects on overall survival were determined (Fig. 4e,f). Combination therapy signi ficantly prolonged survival compared to treatment with vehicle or single-agent therapy for all three T-ALLs (Fig. 4f and Supplementary Table 4). Notably, primary leukemia cells treated with a single GSI agent in vivo had higher H3K27me3 levels, analogous to the lines treated in vitro (Supplementary Fig. 7c). These data support the in vivo relevance of our study and the potential of this combination of targeted therapies to augment current and emerging therapies for T-ALL.
Nature Genetics VOLUME 46 | NUMBER 4 | APRIL 2014
Therapeutic resistance plagued early GSI trials in humans5 and is a major challenge in cancer treatment, pertinent to conventional chemotherapy and targeted therapy25. We have shown that T-ALLs can acquire GSI resistance by a fully reversible epigenetic mechanism. Persister cells rely on alternative signaling pathways to proliferate in the absence of NOTCH1 signaling and appear to arise from rare drug-tolerant cells already present in naive T-ALL populations. Persister cells also exhibit an altered chromatin state and enhanced sensitivity to BRD4 inhibition. Although the chromatin changes are reminiscent of an established model of drug-tolerant lung cancer cells26, the underlying mechanisms appear to be distinct, as we do not observe upregulation of KDM5A expression or sensitivity to histone deacetylase inhibitors (Supplementary Figs. 1f and 2e). The in vivo relevance of our findings is supported by the efficacy of combinatorial therapy with NOTCH and BRD4 inhibitors in patient-derived xenograft models of pediatric T-ALL. The effect of combination therapy in vivo is particularly striking given the short duration of treatment, the fact that treatment was initiated upon reaching a substantial leukemic burden and the refractory nature of the relapsed samples examined. Our study provides a framework for understanding epigenetic alterations and heterogeneity in tumor pathogenesis and suggests that drug resistance may be addressed by combination therapies that incorporate epigenetic modulators. Methods Methods and any associated references are available in the online version of the paper. Accession codes. Gene expression and ChIP-seq data have been deposited in the Gene Expression Omnibus (GEO) under accession GSE54380. Note: Any Supplementary Information and Source Data files are available in the online version of the paper. Acknowledgments We are grateful to T. Look, M. Harris, L. Silverman and S. Sallan for providing the pediatric T-ALL samples. We thank the Flow Cytometry Core of the Harvard Stem Cell Institute and the Center for Regenerative Medicine and Technology for excellent flow sorting. We thank E. Rheinbay, M. Suva, R. Ryan, N. Riggi, A. Goren, O. Ram, J. Wu, L. Pan, W. Pear and M. Rivera for helpful discussions; V. Mootha for critical comments on the manuscript; S. Muller-Knapp and the Structural Genomics Consortium (Oxford, UK) for their assistance with inhibitor experiments; L. Hamm and the Broad RNAi Platform for help with the shRNA screen; R. Issner, X. Zhang, C. Epstein, N. Shoresh, T. Durham and the Broad Genome Sequencing Platform for technical assistance; A. Christie for help with mouse experiments; L. Gaffney for help with illustrations; and O. Weigert for providing the BCL2 ORF. B.K. was supported by a US National Institutes of Health (NIH) T32 training grant (HL007574-30) and by a St. Baldrick’s fellowship. J.E.R. was supported by a US NIH T32 training grant (CA130807) and by a postdoctoral fellowship from the American Cancer Society (125087-PF-13247-01-LIB). H.W. is supported by a US NIH T32 training grant (HL007627). This work was supported by the National Human Genome Research Institute (ENCODE U54 HG004570 to B.E.B.), the NIH Common Fund for Epigenomics (U01 ES017155 to B.E.B.), the NIH/NCI (CA096899 to M.A.K. and 5P01 CA109901-10 to J.E.B.), the Howard Hughes Medical Institute (B.E.B.) and the Starr Cancer Consortium (B.E.B.). The Leukemia & Lymphoma Society Specialized Center of Research Program supports B.E.B., J.C.A., J.E.B., K.S. and A.L.K. AUTHOR CONTRIBUTIONS B.K. and J.E.R. designed and performed experiments and analyzed the data. M.A.K. and B.E.B. designed the experimental strategy and supervised the study and analysis. J.Z. and D.F. carried out computational analyses. B.K., J.E.R., M.A.K. and B.E.B. drafted the manuscript. K.E.W., S.M.G., M.J.C., J.G.L., M.K., H.W., F.P., S.J.S., M.J., D.P., M.J.K. and C.J.O. contributed to experiments and data analysis. L.D.S., M.A.B., D.L.G., A.G., K.S., A.L.K., D.E.R., J.E.B. and J.C.A. provided reagents, contributed to analysis and gave conceptual advice. All authors discussed the results and implications and reviewed the manuscript.
369
letters COMPETING FINANCIAL INTERESTS The authors declare competing financial interests: details are available in the online version of the paper.
1. Weng, A.P. et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 306, 269–271 (2004). 2. Ellisen, L.W. et al. TAN-1, the human homolog of the Drosophila notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 66, 649–661 (1991). 3. Pui, C.H. & Evans, W.E. Treatment of acute lymphoblastic leukemia. N. Engl. J. Med. 354, 166–178 (2006). 4. Rao, S.S. et al. Inhibition of NOTCH signaling by γ secretase inhibitor engages the RB pathway and elicits cell cycle exit in T-cell acute lymphoblastic leukemia cells. Cancer Res. 69, 3060–3068 (2009). 5. Palomero, T. & Ferrando, A. Therapeutic targeting of NOTCH1 signaling in T-cell acute lymphoblastic leukemia. Clin. Lymphoma Myeloma 9 (suppl. 3), S205–S210 (2009). 6. Guruharsha, K.G., Kankel, M.W. & Artavanis-Tsakonas, S. The Notch signalling system: recent insights into the complexity of a conserved pathway. Nat. Rev. Genet. 13, 654–666 (2012). 7. Weng, A.P. et al. c-Myc is an important direct target of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev. 20, 2096–2109 (2006). 8. Palomero, T. et al. NOTCH1 directly regulates c-MYC and activates a feed-forwardloop transcriptional network promoting leukemic cell growth. Proc. Natl. Acad. Sci. USA 103, 18261–18266 (2006). 9. Sharma, V.M. et al. Notch1 contributes to mouse T-cell leukemia by directly inducing the expression of c-myc. Mol. Cell. Biol. 26, 8022–8031 (2006). 10. Gutierrez, A. et al. High frequency of PTEN, PI3K, and AKT abnormalities in T-cell acute lymphoblastic leukemia. Blood 114, 647–650 (2009). 11. Palomero, T. et al. Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T-cell leukemia. Nat. Med. 13, 1203–1210 (2007).
npg
© 2014 Nature America, Inc. All rights reserved.
Reprints and permissions information is available online at http://www.nature.com/ reprints/index.html.
12. Chan, S.M., Weng, A.P., Tibshirani, R., Aster, J.C. & Utz, P.J. Notch signals positively regulate activity of the mTOR pathway in T-cell acute lymphoblastic leukemia. Blood 110, 278–286 (2007). 13. Kalaitzidis, D. et al. mTOR complex 1 plays critical roles in hematopoiesis and Pten-loss-evoked leukemogenesis. Cell Stem Cell 11, 429–439 (2012). 14. Dawson, M.A. & Kouzarides, T. Cancer epigenetics: from mechanism to therapy. Cell 150, 12–27 (2012). 15. Zhao, R., Nakamura, T., Fu, Y., Lazar, Z. & Spector, D.L. Gene bookmarking accelerates the kinetics of post-mitotic transcriptional re-activation. Nat. Cell Biol. 13, 1295–1304 (2011). 16. Blobel, G.A., Kalota, A., Sanchez, P.V. & Carroll, M. Short hairpin RNA screen reveals bromodomain proteins as novel targets in acute myeloid leukemia. Cancer Cell 20, 287–288 (2011). 17. Zuber, J. et al. RNAi screen identifies Brd4 as a therapeutic target in acute myeloid leukaemia. Nature 478, 524–528 (2011). 18. Filippakopoulos, P. et al. Selective inhibition of BET bromodomains. Nature 468, 1067–1073 (2010). 19. Nicodeme, E. et al. Suppression of inflammation by a synthetic histone mimic. Nature 468, 1119–1123 (2010). 20. Zhou, V.W., Goren, A. & Bernstein, B.E. Charting histone modifications and the functional organization of mammalian genomes. Nat. Rev. Genet. 12, 7–18 (2011). 21. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012). 22. Whyte, W.A. et al. Master transcription factors and mediator establish superenhancers at key cell identity genes. Cell 153, 307–319 (2013). 23. Voigt, P. & Reinberg, D. BRD4 jump-starts transcription after mitotic silencing. Genome Biol. 12, 133 (2011). 24. Dawson, M.A. et al. Inhibition of BET recruitment to chromatin as an effective treatment for MLL-fusion leukaemia. Nature 478, 529–533 (2011). 25. Haber, D.A., Gray, N.S. & Baselga, J. The evolving war on cancer. Cell 145, 19–24 (2011). 26. Sharma, S.V. et al. A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell 141, 69–80 (2010).
370
VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
ONLINE METHODS
npg
© 2014 Nature America, Inc. All rights reserved.
Cell culture. The human T cell leukemia cell lines DND-41 and KOPT-K1 were a kind gift of A.T. Look (Dana-Farber Cancer Institute). Cells were grown in RPMI1640 containing 10% FCS at 37 °C with 5% CO2 , tested for mycoplasma at regular intervals and maintained at a density of between 5 × 105 and 2 × 106 cells/ml. Persister cells were established by treating DND-41 or KOPT-K1 cells with 1 µM GSI for at least 7 weeks, replenishing the inhibitor every 3–4 d (Compound E, EMD4 Biosciences). Reversed cells were generated from persister cells by culturing without GSI for a minimum of 2 weeks. Cell numbers were determined using the Nexcelom Bioscience Cellometer Auto T4. Single cells were sorted into 96-well round-bottom plates and continuously treated with 1 µM Compound E or control. The number of clones that grew out after 4–6 weeks of culture was determined using a Zeiss Observer.Z1 with AxioCam MRm with a 2.5× objective imaging system. At least 500 wells were analyzed per condition in 2 independent experiments. Primary T-ALL cells obtained from pediatric patients with T-ALL with institutional review board (IRB) approval were passaged through NOD-scid IL2Rgnull (NSG) mice and maintained in MEMα containing 10% FCS, 10% heatinactivated human AB+ serum, 10 ng/ml human interleukin (IL)-7, 50 ng/ml human stem cell factor (SCF), 20 ng/ml human FLT3 ligand, 20 nM insulin and 100 ng/ml IL-2 at 37 °C with 5% CO2. Cell proliferation and apoptosis assays. Viable T-ALL cell lines were plated in 2–4 replicates in black or white opaque flat-bottom 96-well tissue culture plates. For proliferation assays, cells were titrated to allow log-phase growth for a period of 4 to 9 d before readout (5,000 to 15,000 cells per well). Viable primary T-ALL cells were plated in 2–3 replicates in black opaque flat-bottom 96-well tissue culture plates at 50,000–100,000 cells per well and were cultured for 6 d. Cell proliferation was measured using the CellTiter-Glo Luminescent Cell Viability Assay from Promega as described by the manufacturer. Apoptosis was measured as a function of caspase 3 and caspase 7 cleavage using the Caspase-Glo 3/7 Luminescent Assay from Promega as described by the manufacturer. End-point luminescence was measured on a SpectraMax M5 plate reader (Molecular Diagnostics). Statistical analyses were performed using Student’s t tests, assuming equal variance. A two-sided P value of
npg
© 2014 Nature America, Inc. All rights reserved.
An epigenetic mechanism of resistance to targeted therapy in T cell acute lymphoblastic leukemia Birgit Knoechel1–4,16, Justine E Roderick5,16, Kaylyn E Williamson1,2,6,7, Jiang Zhu1,2,6,7, Jens G Lohr2,8, Matthew J Cotton1,2,6,7, Shawn M Gillespie1,2,6,7, Daniel Fernandez2,9, Manching Ku1,2,6, Hongfang Wang10, Federica Piccioni2, Serena J Silver2, Mohit Jain2,11,12, Daniel Pearson4,13, Michael J Kluk10, Christopher J Ott8, Leonard D Shultz14, Michael A Brehm15, Dale L Greiner15, Alejandro Gutierrez3,4, Kimberly Stegmaier3,4, Andrew L Kung3,4, David E Root2, James E Bradner2,8, Jon C Aster10, Michelle A Kelliher5 & Bradley E Bernstein1,2,6,7 The identification of activating NOTCH1 mutations in T cell acute lymphoblastic leukemia (T-ALL) led to clinical testing of g-secretase inhibitors (GSIs) that prevent NOTCH1 activation1–4. However, responses to these inhibitors have been transient5, suggesting that resistance limits their clinical efficacy. Here we modeled T-ALL resistance, identifying GSI-tolerant ‘persister’ cells that expand in the absence of NOTCH1 signaling. Rare persisters are already present in naive T-ALL populations, and the reversibility of their phenotype suggests an epigenetic mechanism. Relative to GSI-sensitive cells, persister cells activate distinct signaling and transcriptional programs and exhibit chromatin compaction. A knockdown screen identified chromatin regulators essential for persister viability, including BRD4. BRD4 binds enhancers near critical T-ALL genes, including MYC and BCL2. The BRD4 inhibitor JQ1 downregulates expression of these targets and induces growth arrest and apoptosis in persister cells, at doses well tolerated by GSI-sensitive cells. Consistently, the GSI-JQ1 combination was found to be effective against primary human leukemias in vivo. Our findings establish a role for epigenetic heterogeneity in leukemia resistance that may be addressed by incorporating epigenetic modulators in combination therapy. T-ALL is an aggressive malignancy frequently associated with activating mutations in NOTCH1, a critical oncogene in this disease1. GSIs that prevent NOTCH1 cleavage and activation have been tested in clinical trials and mouse models, but responses have been modest and
transient5. To understand the mechanisms by which T-ALL cells overcome NOTCH1 inhibition, we modeled GSI resistance in vitro and in vivo and investigated the functional and molecular characteristics of resistant cells. By chronically exposing NOTCH1-dependent human T-ALL cells to a GSI (compound E) in vitro, we isolated a population of persister cells that tolerated GSI concentrations more than 50-fold higher than the concentrations tolerated by naive cells (Fig. 1a,b and Supplementary Figs. 1a and 2a). To characterize these drug-tolerant cells, we first examined the active intracellular form of NOTCH1 (ICN1)6. ICN1 was present at high levels in naive T-ALL cells but was essentially undetectable in persister cells (Fig. 1c, top and Supplementary Fig. 2b, top). Expression of NOTCH1 target genes, including DTX1 and HES4, was also downregulated in persister cells (Supplementary Figs. 1b,c and 3i). The phenotype was reversible, as persister cells rapidly re-expressed ICN1 and NOTCH1 target genes after removal of the GSI (Fig. 1c and Supplementary Figs. 1b,d and 2b). Re-exposure of these ‘reversed’ cells to GSI led to downregulation of the expression of NOTCH1 target genes and growth arrest, as in the original naive T-ALL population (Supplementary Fig. 1b). This reversibility suggests that the resistance phenotype is mediated epigenetically. Induction of MYC is thought to be a major mechanism whereby constitutive NOTCH1 activation results in leukemic T cell transformation7–9. Although MYC protein levels were dramatically reduced by short-term GSI treatment, moderate levels were maintained in the persister cells, despite continued GSI treatment (Fig. 1c and Supplementary Fig. 2b). Consistently, the strong MYC transcriptional
1Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, USA. 2Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA. 3Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA. 4Division of Hematology/Oncology, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, USA. 5Department of Cancer Biology, University of Massachusetts Medical School, Worcester, Massachusetts, USA. 6Center for Cancer Research, Massachusetts General Hospital, Boston, Massachusetts, USA. 7Howard Hughes Medical Institute, Chevy Chase, Maryland, USA. 8Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA. 9Biostatistics Graduate Program, Harvard University, Cambridge, Massachusetts, USA. 10Department of Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, USA. 11Center for Human Genetic Research, Massachusetts General Hospital, Boston, Massachusetts, USA. 12Department of Molecular Biology, Massachusetts General Hospital, Boston, Massachusetts, USA. 13Biological and Biomedical Sciences Graduate Program, Harvard Medical School, Boston, Massachusetts, USA. 14Jackson Laboratory, Bar Harbor, Maine, USA. 15Program in Molecular Medicine, University of Massachusetts Medical School, Worcester, Massachusetts, USA. 16These authors contributed equally to this work. Correspondence should be addressed to B.E.B. ([email protected]) or M.A.K. ([email protected]).
Received 12 August 2013; accepted 6 February 2014; published online 2 March 2014; doi:10.1038/ng.2913
364
VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
letters
1 0 Naive
Persister
d
0.4
mTOR
MYC
0.2
Tubulin
0
Tubulin
P
Naive Persister
1.2 1.0 0.8 0.6 0.4 0.2 0
N ST
0
0.625 1.25 2.5 5 Rapamycin (nM)
10
Figure 1 A subpopulation of drug-tolerant T-ALL cells mediates Persister Naive resistance to NOTCH1 inhibition. 1.0 Reversed 1.0 Rapamycin pretreated 1.0 Naive (a) Bar plot comparing the expansion 0 0 0.8 0.8 0.8 rates of native T-ALL cells (grown in the absence 0.6 MYC 0.6 0.6 of GSI) and persister cells grown 0.4 0.4 0.4 continuously in 1 µM GSI (two Tubulin replicates; error bars, s.d.). 0.2 0.2 0.2 *** *** (b) Proliferation of naive and *** 0 0 0 persister cells treated with Control Treated Control Treated Control Treated the indicated doses of NOTCH inhibitor for 6 d (two replicates; error bars, s.d.). (c) Protein blot showing ICN1 and MYC levels in naive cells (N), cells with short-term treatment (ST; 5 d of 1 µM GSI), persister cells in 1 µM GSI (P), reversed persister cells removed from GSI for 2 weeks (Rev) and reversed cells re-exposed to GSI (Rev tx; 5 d). Reactivation of NOTCH1 signaling in reversed cells suggests that the persister phenotype is reversible. (d) Protein blot showing phosphorylated mTOR (p2481), a marker of activated mTOR signaling, total mTOR and tubulin in naive (N) and persister (P) cells. (e) Proliferation of naive and persister cells treated with the indicated concentrations of rapamycin for 6 d (four replicates; error bars, s.d.). (f) Protein blots showing MYC expression in naive and persister cells after 3 d of treatment with 2 µM AKT inhibitor MK-2206 (AKTi) or 10 nM mTOR inhibitor rapamycin (Rapa). (Data shown are for DND-41 cells; data for KOPT-K1 cells are shown in Supplementary Fig. 2). (g) Rare drug-tolerant cells pre-exist in naive T-ALL populations. Single cells from a naive DND-41 population, a reversed population (reversed for 1 month) and a naive population pretreated with 10 nM rapamycin for 3 d were sorted into individual wells of 96-well plates and cultured in 1 µM GSI (treated) or without drug (control) for 4–6 weeks. Bar plots show the fraction of single cells that form colonies (sorted single cells per condition: naïve, n = 936; reversed, n = 936; rapamycin pretreated, n = 858; pooled from 2 independent experiments; ***P < 0.001; error bars, s.d.). These data suggest that GSI tolerance is already evident in a small fraction of naive T-ALL cells and is reversible.
signature of naive T-ALL cells was reduced in persister cells (Supplementary Table 1). Conversely, gene signatures associated with mitogen-activated protein kinase (MAPK), c-Jun N-terminal kinase (JNK), phosphoinositide 3-kinase (PI3K) and mTOR signaling were enhanced (Supplementary Table 1). Genetic alterations of PTEN and consequent AKT pathway activation have been associated with GSI resistance10,11. Although the T-ALL cells modeled here did not harbor PTEN mutations11, persister cells had higher levels of phosphorylated PTEN and were sensitive to AKT inhibitors (Supplementary Figs. 1e and 2d). mTOR activation has also been observed in T-ALL 12,13. Persister cells had higher levels of p2481 (phosphorylated mTOR) and were sensitive to the mTOR inhibitor rapamycin (Fig. 1d,e and Supplementary Fig. 2c). Inhibition of mTOR markedly reduced MYC protein levels in persister cells but not in naive T-ALL cells (Fig. 1f). In contrast, persister cells did not exhibit enhanced sensitivity to doxo rubicin or vorinostat (Supplementary Figs. 1f and 2e). These data suggest that rewired signaling programs maintain MYC expression and cell proliferation in the absence of NOTCH1 activity. Next, we sought to determine whether resistance is induced by GSI treatment or whether, alternatively, a fraction of unexposed cells is already drug tolerant. In support of the latter hypothesis, when we isolated single cells from a naive T-ALL population, we found that ~4% expanded in the presence of GSI (Fig. 1g). In contrast, the remaining cells failed to grow, even after extended periods of time. An analogous analysis of reversed cells, which were exposed to GSI but then expanded in the absence of drug for 1 month, again showed a small fraction of GSI-tolerant single cells (~9%). We hypothesized that pre-existing GSI-tolerant cells share phenotypic characteristics with persister cells, such as mTOR dependency. In support of this Nature Genetics VOLUME 46 | NUMBER 4 | APRIL 2014
Fraction of single-cell clones
Fraction of single-cell clones
Fraction of single-cell clones
AK
AK
Ti R ap a
g
Ti R ap a
f
© 2014 Nature America, Inc. All rights reserved.
p2481
P R e R v ev tx
0.6
GSI (µM)
npg
N
Tubulin
0.8
e
ICN1
P R e R v ev tx
2
1.0
c
N ST
3
1.2
50 10 0
Cell survival normalized to control (CellTiter-Glo)
Fold expansion in 3 d
4
Naive Persister
Cell survival normalized to control (CellTiter-Glo)
b
0 0. 4 0. 8 1. 5 3. 1 6. 2 12 .5 25
a
idea, pretreatment with rapamycin reduced the frequency of GSItolerant single cells by ~5-fold (Fig. 1g). These data suggest that GSI resistance depends on a pre-existing subpopulation of T-ALL cells with distinct signaling programs that readily expand in the absence of NOTCH1 signaling. The reversibility of the phenotype suggests that this heterogeneity is mediated by epigenetic rather than genetic alterations. In addition to altered signaling, persister cells exhibited morphologic changes, including decreased cell and nuclear sizes (Fig. 2a,b and Supplementary Fig. 3a,b). We postulated that these changes might reflect global chromatin changes associated with NOTCH1 inactivity. We found that the persister cells had higher global levels of repressive histone modifications and heterochromatin protein 1 (HP1) (Fig. 2c,d and Supplementary Fig. 3c,e). Chromatin in persister cells was also relatively inaccessible to micrococcal nuclease (MNase) digestion (Supplementary Fig. 3g), consistent with a compact state. Chromatin compaction likely reflects the absence of NOTCH1 activity, as it was also evident in cells with short-term GSI treatment and in T-ALL cells in which NOTCH1 signaling was inhibited by dominant-negative mastermind-like 1 (DN-MAML1) (Fig. 2c and Supplementary Figs. 1c and 3c,e,i). We postulated that the altered chromatin state in persister cells might uncover new susceptibilities that could be targeted by emerging epigenetic therapies. To test this hypothesis, we designed a lentiviral short hairpin RNA (shRNA) knockdown screen that targeted ~350 chromatin regulators with an average of 5 independent hairpins per gene (Online Methods). We identified ~15 genes for which knockdown compromised the survival of naive and persister cells, which were thus presumed to be generally required in T-ALL 365
npg
H3K9me3/total H3
H3K27me3/total H3
600,000 400,000 200,000
0.050
** 0.025
0
N ST P BRD4 Tubulin
1.0
0.20
tx
1
0
0.25
ut
0.5
JQ
0.0625 0.125 0.25 JQ1 (µM)
o
0
N
0
ho
SND1 SS18 STAG3L1 SUDS3
0.01
800,000
0.075
as
PRMT2 EZH2 SUV39H1 PRMT5 PRMT7 KDM5D MLL5 MINA MORF4L1
0.02
1,000,000
w
te
ai
is rs
Pe
N
r
0
ARID3B BRD1 BRD4 CD2BP2 CHD1 CHD8 CHRAC1
0.03
1,200,000
0.100
1
0.05
**
0
e-
Preferential dependency persister
0.04
5
0
pr
0.10
0.05
0.5
h
1,400,000
10
1
0.15
2.5×
15
JQ
0.20
0.06
1.0
**
20
JQ
0.25
HMGXB4 PRDM4 HMG20B PRMT3 MBD6 PRMT8 SMOX SPTY2D1
***
1.5
Naive Persister
1,600,000
Sh N tre or aive at t-te m r Pe en m rs t is te r
**
HDAC2 HDAC4 HDAC5 HDAC6 HDAC9
BRD4 mRNA normalized to B2M
0.30
ve
CBX3 mRNA normalized to B2M
© 2014 Nature America, Inc. All rights reserved.
Preferential dependency naive ARID5A BAZ1B CHAF1B CHMP4A CHD5 CHD7 EHMT2
te r
g 0.07
2.0
Sh Na i tre ort ve at -te m rm e Pe nt rs is te r
f
*
25
Fraction of single cell clones normalized to untreated
e
500 300
30
**
N Sh aiv tre ort e at -te m rm e Pe nt rs is te r
d
600
is
0 200 400 600 800 1,000 FSC
700
e
0
800
ai v
20
350
*
2.5
Pe rs
40
400
900
N
60
Nuclear area (pixels)
80
Persister
Cell survival (CellTiter-Glo)
Naive Persister
c 3.0 Naive
Fraction of apoptotic cells (Caspase-Glo/CellTiter-Glo)
Percentage of maximum
100
b
450
N ai Pe ve rs is te R r ev er se R d et re at ed
a
FSC (median fluorescence intensity)
letters
0.15
Figure 2 Drug-tolerant T-ALL cells adopt an altered 0.10 chromatin state and are BRD4 dependent. (a) Left, forward scatter (FSC) distribution of naive and persister cells determined by flow cytometry; right, bar 0.05 plot of FSC indicating the size distributions of naive (blue), persister (red) and 0 reversed (green) cells and reversed cells re-exposed to GSI for 5 d (retreated; orange). 0 0.0625 0.125 0.25 0.5 1.0 (b) Sizes of naive and persister cell nuclei are shown by 4,6-diamidino-2-phenylindole JQ1 (µM) (DAPI) staining (left; scale bars, 50 µm) and quantified in the box plots (right; naive, n = 585; persister, n = 496; P < 1 × 10−4; boxes indicate 25th, 50th and 75th percentiles, whiskers represent 5th and 95th percentiles). (c) Bar plots indicating relative levels of repressive histone modifications per ELISA of bulk histones from naive cells undergoing short-term treatment (3 d, 1 µM GSI) and persister cells (two (technical) replicates; error bars, s.d.; *P < 0.05, **P < 0.01). (d) CBX3 mRNA (encoding HP1γ) expression shown for naive and persister T-ALL cells (two replicates; error bars, s.d.; **P < 0.01). (e) The shRNA screen identified chromatin regulators preferentially required for naive or persister cell survival. Top hits for each cell state are indicated. (f) BRD4 expression is shown for naive cells (N), cells undergoing short-term treatment (ST; 5 d) and persister cells (P) (top, mRNA expression; bottom, protein expression; three (technical) replicates; error bars, s.d.; **P < 0.01, ***P < 0.001). (Data shown are for DND-41 cells; data for KOPT-K1 cells are shown in Supplementary Fig. 3). (g) Proliferation (6 d of treatment; four replicates) and rate of apoptosis (4 d of treatment; three replicates) are shown for naive and persister cells treated with the indicated doses of JQ1 (error bars, s.d.). (h) Naive DND-41 populations were pretreated with 0.5 µM JQ1 for 4 d (JQ1 pre-tx), pretreated with 0.5 µM JQ1 for 4 d but then cultured for 24 h without JQ1 (JQ1 washout) or never exposed to JQ1 (no JQ1). Single cells from each of these populations were sorted into individual wells of 96-well plates and cultured with GSI (1 µM) or without GSI (control) for 6 weeks. The bar plot indicates the fraction of single cells that form colonies in GSI relative to control cells (sorted single cells per condition: n = 936; data pooled from 2 independent experiments; error bars, s.d. ; **P < 0.01). Pretreatment with JQ1 eliminates pre-existing GSI-tolerant cells from naive T-ALL populations, but this effect is reversed by 24 h of JQ1 washout.
(Supplementary Table 2). We also identified genes preferentially required for the survival of either naive or persister cells. For example, naive cells were preferentially dependent on several histone deacetylases, whereas persister cells were more dependent on certain arginine methyltransferases and histone methyltransferases (Fig. 2e and Supplementary Table 2). We identified BRD4 as a top hit in the screen, with three shRNAs noticeably reducing the proliferation of persister cells without affecting the proliferation of the naive population (Supplementary Fig. 4a,b and Supplementary Table 2). BRD4 is a member of the BET family of bromodomain proteins that bind acetylated histones14,15. BRD4 has been implicated in several malignancies, including acute myeloid leukemia and lymphoma16,17. BRD4 expression (mRNA and protein) increased with NOTCH1 inhibition and was higher in persister cells (Fig. 2f and Supplementary Fig. 3f). 366
This BRD4 dependency is of particular interest given the development of small molecule BET inhibitors, such as JQ1 (refs. 18,19). Indeed, we found that persister cells exhibited ~5-fold enhanced sensitivity to JQ1 and underwent proliferation arrest and apoptosis in response to JQ1 concentrations that were well tolerated by naive T-ALL cells (Fig. 2g and Supplementary Figs. 2e and 4c). We therefore considered whether the preexisting GSI-tolerant cells, identified in naive T-ALL populations by single-cell analysis (Fig. 1g), might also be BRD4 dependent. Indeed, when we pretreated naive T-ALL cells with JQ1 for 4 d and then removed the JQ1 immediately before isolating individual cells into 96-well plates, the frequency of GSI-tolerant single cells was reduced by ~20-fold (Fig. 2h). However, when we pretreated naive cells but then removed JQ1 24 h before isolating individual cells, we were able to detect GSI-tolerant VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
letters d
15 H3K36me3
80,000 60,000
RPTOR CLEC16A CDK6 ETV6 BCL2L1 BCL2
40,000 20,000 0 0
200
600
1,000
Genes ranked by BRD4 signal
BRD4 15 H3K36me3
e
Persister
H3K36me3 BRD4
hg19
hg19
10 kb 50 50 H3K36me3 BRD4 50
JQ1 (µM) 0 0.25 0.50
0 0.25 0.50
BCL2
Naive
Persister
H3K36me3 BRD4
200 kb
hg19
MYC
20 80
Tubulin
20
H3K36me3
Naive
3 2 1 Persister
g EV
50,000
BCL2 ORF
**
40,000 30,000
***
20,000
**
10,000
**
0
Persister
0
ETV6
h
0.0625 0.125 0.25 JQ1 (µM)
0.5
Figure 3 BRD4 binds enhancers near developmental, cell cycle and prosurvival genes in T-ALL. 100,000 EV (a) Heat map showing enrichment signals for BRD4, H3K27ac and H3K4me1 over 29,978 MYC ORF H3K4me1-marked distal sites in persister cells (rows represent 10-kb regions centered 80,000 on H3K4me1 peaks ranked by overall signal intensities for BRD4 and H3K27ac). (b) BRD4 peaks 60,000 were ranked by the signal intensity of the area under the curve (AUC) in persister cells, averaging BRD4 signal intensities of DND-41 and KOPT-K1 persister cells to identify candidate superenhancers. 40,000 The plot depicts top-ranked sites linked to an active gene in T-ALL (Online Methods). (c) Tracks * showing BRD4 binding and H3K36me3 enrichment (marking transcribed regions) over the CDK6 and 20,000 ETV6 loci, both of which contain BRD4-bound superenhancers. (d,e) Left, tracks showing BRD4 binding and H3K36me3 enrichment across the BCL2 (d) and MYC (e) loci in naive and persister cells. 0 Right, plots showing BRD4 enrichment as measured by ChIP and quantitative PCR (qPCR) over putative 0 0.0625 0.125 0.25 0.5 JQ1 (µM) regulatory elements in the BCL2 and MYC loci in naive and persister cells (two replicates; error bars, s.d.; BC, background control). (f) Protein blots showing BCL2 and MYC expression in naive and persister cells after 4 d of treatment with the indicated JQ1 doses. (Data shown are for KOPT-K1 cells; data for DND-41 cells are shown in Supplementary Fig. 5). (g) Plot showing relative proliferation of DND-41 persister cells transfected with BCL2 expression vector (BCL2 ORF) or empty vector control (EV) after 6 d of treatment with the indicated JQ1 doses (three replicates; error bars, s.d.; **P < 0.01, ***P < 0.001). (h) Plot showing relative proliferation of DND-41 persister cells transfected with MYC expression vector (MYC ORF) or empty vector control (EV) after 6 d of treatment with the indicated JQ1 doses (two replicates; error bars, s.d.; *P < 0.05). Cell survival (CellTiter-Glo)
© 2014 Nature America, Inc. All rights reserved.
npg
80 BRD4
4
Naive
30 CDK6
5
0
MYC
H3K36me3
5
6 MYC locus
H3K36me3
f
10
Naive Persister
BRD4
30 80
15
BCL2
Cell survival (CellTiter-Glo)
BRD4 Naive
200 kb
Persister
80
BCL2 locus
20
0 BCL2
50
c
25
BRD4 binding enrichment/BC
100,000
Persister
70
Naive
BRD4 signal (AUC)
H3K4me1 sites (n = 29,978)
hg19
200 kb 70 BRD4
BRD4 binding enrichment/BC
MSI2
120,000
Naive
b
BR D H 4 3K 27 a H 3K c 4m e1
a
clones in a proportion similar to that found in control naive T-ALL cells. Thus, pre-existing GSI-tolerant cells in naive T-ALL populations are highly BRD4 dependent, like persister cells, and exist in dynamic equilibrium with the more prevalent GSI-sensitive cells. To investigate its regulatory functions, we mapped BRD4 binding in naive and persister T-ALL populations by chromatin immunoprecipitation and sequencing (ChIP-seq). We also mapped histone modifications associated with promoters (trimethylation of histone H3 at lysine 4, H3K4me3), transcripts (trimethylation of histone H3 at lysine 36, H3K36me3), enhancers (monomethylation of histone H3 at lysine 4, H3K4me1 and acetylation of histone H3 at lysine 27, H3K27ac) and Polycomb-repressed loci (trimethylation of histone H3 at lysine 27, H3K27me3)20,21. In both cell states, BRD4 bound promoters (~30% of sites) and putative enhancers enriched for H3K4me1 and H3K27ac marks (~60%) (Fig. 3a,b and Supplementary Fig. 5a). We collated the largest BRD4 binding sites in persister cells, as Nature Genetics VOLUME 46 | NUMBER 4 | APRIL 2014
these might reflect ‘superenhancers’ with critical functions in cell type–specific gene regulation22. Proximal genes encoded many known T-ALL regulators, including the transcription factor ETV6, the cell cycle regulator CDK6, the signaling molecule RPTOR and the prosurvival protein BCL2 (Fig. 3b–d, Supplementary Fig. 5b,d and Supplementary Table 3). Although most large BRD4 peaks were shared by the populations, some were specific to either naive cells (for example, those in NOTCH1 target loci) or persister cells (Supplementary Fig. 5c). Thus, BRD4 binds and may sustain the activity of the regulatory elements and target genes required for the proliferation of T-ALL cells. We next considered the molecular basis for the preferential BRD4 dependency in persister cells. The relatively consistent BRD4 binding patterns in naive and persister cells suggest that BRD4 relocalization is unlikely to explain the different dependencies of these cells. However, persister cells exhibited an altered chromatin environment with greater 367
letters tions in potential regulatory regions (Supplementary Fig. 3h). BRD4 is believed to have an important role as a ‘bookmark’ of active regulatory elements, maintaining their chromatin state as cells progress through mitosis15,23. Thus, we suggest that generalized chromatin repression in
ICN1
Cell survival (CellTiter-Glo)
MYC
0.0015
0.0010
***
0.0005
0
V
100,000
0.020
0.015
0.010
0.005
0
LT
V
80,000 70,000 60,000
**
50,000 40,000
**
30,000
**
20,000 10,000 0
e NSG mice
25% leukemic blasts in the periphery
DBZ + JQ1
0.05 0.04 0.03 0.02 0.01 0
80,000
V
**
20,000 Veh GSI
100
4,000 3,000 2,000 1,000 0
V
LT
T-ALL-x-14
9,000
40,000
f Monitor for disease latency
**
*
5,000
LT
60,000
0
Administer treatment for 3 weeks
JQ1
0.06
*
100,000
GSI JQ1 GSI + JQ1
DBZ
0.07
T-ALL-x-11
Vehicle
1 × 106 cells
*
0.08
120,000
Percent survival
Primary leukemia
6,000
0.09
LT
140,000
T-ALL-x-9
90,000
0
8,000 7,000 6,000 5,000 4,000 3,000
**
2,000 1,000 0
JQ1 GSI + JQ1
0
GSI
JQ1 GSI + JQ1
T-ALL-x-9
80 60 40
368
Percent survival
20 Figure 4 Combination therapy targeting NOTCH1 and BRD4 in primary T-ALL. (a) NSG mice were 0 injected with luciferase-expressing KOPT-K1 T-ALL 0 20 40 60 80 100 120 140 160 cells, and leukemic mice were treated with Treatment NOTCH inhibitor DBZ for 5 d (ST; three doses), Days after transplantation DBZ for 3 weeks (LT; dosing every other day) or T-ALL-x-11 T-ALL-x-14 vehicle (V) (five mice per group). DBZ-treated 100 100 Vehicle mice were sacrificed when bioluminescence DBZ had plateaued and then significantly increased 80 80 JQ1 (Online Methods and Supplementary Fig. 7a). DBZ + JQ1 60 60 Images show hematoxylin and eosin (H&E) staining and immunohistochemistry for ICN1 40 40 and MYC in bone marrow from the respective leukemic mice. Scale bar, 100 µm. (b) NOTCH1 20 20 target gene expression shown for leukemia cells 0 0 sorted from the spleen of mice treated with vehicle (V) 0 20 40 60 80 100 130 150 170 190 0 10 110 or with long-term treatment (LT). Data points reflect Treatment Treatment averages for 2 mice (2 replicates; error bars, s.d.; Days after transplantation Days after transplantation ***P < 0.001). (c) Left, CBX3 expression shown for leukemia cells as in b (two replicates; error bars, s.d.; *P < 0.05). Right, intracellular BCL2 expression measured by flow cytometry shown for leukemia cells as in b (two replicates; error bars, s.d.; *P < 0.05). (d) Proliferation of three primary T-ALLs grown in vitro for 6 d in the presence of 1 µM GSI, 0.25 µM JQ1 or both compounds (T-ALL-x-9 and T-ALL-x-14: two replicates; T-ALL-x-11: three replicates; error bars, s.d.; *P < 0.05, **P < 0.01). (e) Experimental design for primary T-ALL xenotransplantation trials. Primary T-ALL samples from three primary pediatric T-ALLs (T-ALL-x-9, T-ALL-x-11 and T-ALL-x-14; Supplementary Table 4) were transplanted into NSG mice. Once leukemic burden reached 25% in peripheral blood by human CD45 staining, mice were randomized into four treatment groups (vehicle, JQ1, DBZ or DBZ-JQ1 combination). Mice were treated for three consecutive weeks and were then monitored for disease and sacrificed when they became moribund. (f) KaplanMeier survival curves for samples T-ALL-x-9, T-ALL-x-11 and T-ALL-x-14 (P < 0.01 as assessed by the log-rank test; for details see Supplementary Table 4; T-ALL-x-9: vehicle n = 5, DBZ n = 5, JQ1 n = 5, DBZ and JQ1 n = 4; T-ALL-x-11: vehicle n = 4, DBZ n = 5, JQ1 n = 5, DBZ and JQ1 n = 4; T-ALL-x-14: vehicle n = 5, DBZ n = 5, JQ1 n = 5, DBZ and JQ1 n = 4). Combination treatment significantly prolonged survival compared to single-agent treatment and treatment with vehicle in all three trials. Percent survival
npg
© 2014 Nature America, Inc. All rights reserved.
d
0.0020
c
0.025
BCL2 (mean fluorescence intensity)
H&E
0.0025
Cell survival (CellTiter-Glo)
b
LT
Cell survival (CellTiter-Glo)
ST
HES4 mRNA expression normalized to B2M
V
DTX1 mRNA expression normalized to B2M
a
CBX3 mRNA epxression normalized to B2M
compaction, higher levels of repressive histone modifications and lower levels of the enhancer-associated H3K27ac mark (Fig. 2c and Supplementary Fig. 3c–e,g). Consistently, ChIP-seq data showed that persister cells had modestly higher levels of repressive histone modifica-
VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
npg
© 2014 Nature America, Inc. All rights reserved.
letters persister cells renders enhancers particularly dependent on BRD4 for their epigenetic maintenance (Supplementary Fig. 6). We next focused on individual genes that might account for BRD4 dependency. BCL2 is an established BRD4-dependent gene in mixed-lineage leukemia 24. BCL2 was near a top-ranked BRD4 binding site and was expressed at higher levels in persister cells (Fig. 3b,d,f and Supplementary Figs. 3i and 5d). Treatment with JQ1 markedly reduced BCL2 protein expression in persister cells but had little effect on its expression in naive cells (Fig. 3f). Downregulation of BCL2 expression appears to contribute to the reduced survival of JQ1-treated persister cells, as exogenous BCL2 overexpression partially rescued these cells from JQ1 treatment (Figs. 2g and 3g). The MYC oncogene is an established driver in T-ALL and represents a canonical BRD4 target18. A large enhancer in the MYC locus was strongly enriched for BRD4 binding (Fig. 3e and Supplementary Fig. 5f). Treatment with JQ1 noticeably reduced MYC protein expression in persister cells at doses that did not alter MYC expression in naive cells (Fig. 3f). This effect could partially be rescued by MYC overexpression (Fig. 3h). Hence, compromised survival of JQ1-treated persister cells depends on the downregulation of BCL2 and MYC. To investigate the in vivo relevance of GSI resistance and associated epigenetic changes, we injected luciferase-expressing KOPT-K1 TALL cells orthotopically into NOD-scid IL2Rg null (NSG) mice and followed bioluminescence over time. GSI resistance developed rapidly in vivo after a short period of slowed leukemic growth (Supplementary Fig. 7a). ICN1 levels were drastically reduced in the bone marrow of GSI-treated mice, and expression of NOTCH1 target genes was downregulated in the leukemia cells, indicating that resistance is not due to NOTCH1 reactivation (Fig. 4a,b). These in vivo persister cells also shared other phenotypic characteristics with their in vitro counterparts, including reactivation of MYC expression and increased levels of CBX3 mRNA (encoding HP1γ) and BCL2 (Fig. 4a,c). We hypothesized that combined inhibition of NOTCH1 and BRD4 might be an effective treatment for T-ALL, including relapsed or induction failure disease, where conventional therapy has failed. We therefore examined the efficacy of GSI-JQ1 combination therapy in primary T-ALLs in vitro and in vivo. We examined three tumors from pediatric patients collected at diagnosis or upon relapse (Supplementary Fig. 7b and Supplementary Table 4), including a sample isolated from an individual with relapsed early thymic progenitor (ETP) T-ALL, who failed to respond to cytotoxic therapy. First, we examined the sensitivity of these leukemias to GSI, JQ1 or combined treatment in vitro. We found that the combination treatment was significantly more effective at inhibiting cell growth and survival than single-agent treatment (Fig. 4d). To examine effects in vivo, we engrafted NSG mice with the primary T-ALLs and initiated treatment when a leukemic burden of ~25% in peripheral blood was reached. Randomized groups were treated with vehicle, NOTCH inhibitor (DBZ), BRD4 inhibitor (JQ1) or the DBZ-JQ1 combination for 3 weeks, and the effects on overall survival were determined (Fig. 4e,f). Combination therapy signi ficantly prolonged survival compared to treatment with vehicle or single-agent therapy for all three T-ALLs (Fig. 4f and Supplementary Table 4). Notably, primary leukemia cells treated with a single GSI agent in vivo had higher H3K27me3 levels, analogous to the lines treated in vitro (Supplementary Fig. 7c). These data support the in vivo relevance of our study and the potential of this combination of targeted therapies to augment current and emerging therapies for T-ALL.
Nature Genetics VOLUME 46 | NUMBER 4 | APRIL 2014
Therapeutic resistance plagued early GSI trials in humans5 and is a major challenge in cancer treatment, pertinent to conventional chemotherapy and targeted therapy25. We have shown that T-ALLs can acquire GSI resistance by a fully reversible epigenetic mechanism. Persister cells rely on alternative signaling pathways to proliferate in the absence of NOTCH1 signaling and appear to arise from rare drug-tolerant cells already present in naive T-ALL populations. Persister cells also exhibit an altered chromatin state and enhanced sensitivity to BRD4 inhibition. Although the chromatin changes are reminiscent of an established model of drug-tolerant lung cancer cells26, the underlying mechanisms appear to be distinct, as we do not observe upregulation of KDM5A expression or sensitivity to histone deacetylase inhibitors (Supplementary Figs. 1f and 2e). The in vivo relevance of our findings is supported by the efficacy of combinatorial therapy with NOTCH and BRD4 inhibitors in patient-derived xenograft models of pediatric T-ALL. The effect of combination therapy in vivo is particularly striking given the short duration of treatment, the fact that treatment was initiated upon reaching a substantial leukemic burden and the refractory nature of the relapsed samples examined. Our study provides a framework for understanding epigenetic alterations and heterogeneity in tumor pathogenesis and suggests that drug resistance may be addressed by combination therapies that incorporate epigenetic modulators. Methods Methods and any associated references are available in the online version of the paper. Accession codes. Gene expression and ChIP-seq data have been deposited in the Gene Expression Omnibus (GEO) under accession GSE54380. Note: Any Supplementary Information and Source Data files are available in the online version of the paper. Acknowledgments We are grateful to T. Look, M. Harris, L. Silverman and S. Sallan for providing the pediatric T-ALL samples. We thank the Flow Cytometry Core of the Harvard Stem Cell Institute and the Center for Regenerative Medicine and Technology for excellent flow sorting. We thank E. Rheinbay, M. Suva, R. Ryan, N. Riggi, A. Goren, O. Ram, J. Wu, L. Pan, W. Pear and M. Rivera for helpful discussions; V. Mootha for critical comments on the manuscript; S. Muller-Knapp and the Structural Genomics Consortium (Oxford, UK) for their assistance with inhibitor experiments; L. Hamm and the Broad RNAi Platform for help with the shRNA screen; R. Issner, X. Zhang, C. Epstein, N. Shoresh, T. Durham and the Broad Genome Sequencing Platform for technical assistance; A. Christie for help with mouse experiments; L. Gaffney for help with illustrations; and O. Weigert for providing the BCL2 ORF. B.K. was supported by a US National Institutes of Health (NIH) T32 training grant (HL007574-30) and by a St. Baldrick’s fellowship. J.E.R. was supported by a US NIH T32 training grant (CA130807) and by a postdoctoral fellowship from the American Cancer Society (125087-PF-13247-01-LIB). H.W. is supported by a US NIH T32 training grant (HL007627). This work was supported by the National Human Genome Research Institute (ENCODE U54 HG004570 to B.E.B.), the NIH Common Fund for Epigenomics (U01 ES017155 to B.E.B.), the NIH/NCI (CA096899 to M.A.K. and 5P01 CA109901-10 to J.E.B.), the Howard Hughes Medical Institute (B.E.B.) and the Starr Cancer Consortium (B.E.B.). The Leukemia & Lymphoma Society Specialized Center of Research Program supports B.E.B., J.C.A., J.E.B., K.S. and A.L.K. AUTHOR CONTRIBUTIONS B.K. and J.E.R. designed and performed experiments and analyzed the data. M.A.K. and B.E.B. designed the experimental strategy and supervised the study and analysis. J.Z. and D.F. carried out computational analyses. B.K., J.E.R., M.A.K. and B.E.B. drafted the manuscript. K.E.W., S.M.G., M.J.C., J.G.L., M.K., H.W., F.P., S.J.S., M.J., D.P., M.J.K. and C.J.O. contributed to experiments and data analysis. L.D.S., M.A.B., D.L.G., A.G., K.S., A.L.K., D.E.R., J.E.B. and J.C.A. provided reagents, contributed to analysis and gave conceptual advice. All authors discussed the results and implications and reviewed the manuscript.
369
letters COMPETING FINANCIAL INTERESTS The authors declare competing financial interests: details are available in the online version of the paper.
1. Weng, A.P. et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 306, 269–271 (2004). 2. Ellisen, L.W. et al. TAN-1, the human homolog of the Drosophila notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 66, 649–661 (1991). 3. Pui, C.H. & Evans, W.E. Treatment of acute lymphoblastic leukemia. N. Engl. J. Med. 354, 166–178 (2006). 4. Rao, S.S. et al. Inhibition of NOTCH signaling by γ secretase inhibitor engages the RB pathway and elicits cell cycle exit in T-cell acute lymphoblastic leukemia cells. Cancer Res. 69, 3060–3068 (2009). 5. Palomero, T. & Ferrando, A. Therapeutic targeting of NOTCH1 signaling in T-cell acute lymphoblastic leukemia. Clin. Lymphoma Myeloma 9 (suppl. 3), S205–S210 (2009). 6. Guruharsha, K.G., Kankel, M.W. & Artavanis-Tsakonas, S. The Notch signalling system: recent insights into the complexity of a conserved pathway. Nat. Rev. Genet. 13, 654–666 (2012). 7. Weng, A.P. et al. c-Myc is an important direct target of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev. 20, 2096–2109 (2006). 8. Palomero, T. et al. NOTCH1 directly regulates c-MYC and activates a feed-forwardloop transcriptional network promoting leukemic cell growth. Proc. Natl. Acad. Sci. USA 103, 18261–18266 (2006). 9. Sharma, V.M. et al. Notch1 contributes to mouse T-cell leukemia by directly inducing the expression of c-myc. Mol. Cell. Biol. 26, 8022–8031 (2006). 10. Gutierrez, A. et al. High frequency of PTEN, PI3K, and AKT abnormalities in T-cell acute lymphoblastic leukemia. Blood 114, 647–650 (2009). 11. Palomero, T. et al. Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T-cell leukemia. Nat. Med. 13, 1203–1210 (2007).
npg
© 2014 Nature America, Inc. All rights reserved.
Reprints and permissions information is available online at http://www.nature.com/ reprints/index.html.
12. Chan, S.M., Weng, A.P., Tibshirani, R., Aster, J.C. & Utz, P.J. Notch signals positively regulate activity of the mTOR pathway in T-cell acute lymphoblastic leukemia. Blood 110, 278–286 (2007). 13. Kalaitzidis, D. et al. mTOR complex 1 plays critical roles in hematopoiesis and Pten-loss-evoked leukemogenesis. Cell Stem Cell 11, 429–439 (2012). 14. Dawson, M.A. & Kouzarides, T. Cancer epigenetics: from mechanism to therapy. Cell 150, 12–27 (2012). 15. Zhao, R., Nakamura, T., Fu, Y., Lazar, Z. & Spector, D.L. Gene bookmarking accelerates the kinetics of post-mitotic transcriptional re-activation. Nat. Cell Biol. 13, 1295–1304 (2011). 16. Blobel, G.A., Kalota, A., Sanchez, P.V. & Carroll, M. Short hairpin RNA screen reveals bromodomain proteins as novel targets in acute myeloid leukemia. Cancer Cell 20, 287–288 (2011). 17. Zuber, J. et al. RNAi screen identifies Brd4 as a therapeutic target in acute myeloid leukaemia. Nature 478, 524–528 (2011). 18. Filippakopoulos, P. et al. Selective inhibition of BET bromodomains. Nature 468, 1067–1073 (2010). 19. Nicodeme, E. et al. Suppression of inflammation by a synthetic histone mimic. Nature 468, 1119–1123 (2010). 20. Zhou, V.W., Goren, A. & Bernstein, B.E. Charting histone modifications and the functional organization of mammalian genomes. Nat. Rev. Genet. 12, 7–18 (2011). 21. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012). 22. Whyte, W.A. et al. Master transcription factors and mediator establish superenhancers at key cell identity genes. Cell 153, 307–319 (2013). 23. Voigt, P. & Reinberg, D. BRD4 jump-starts transcription after mitotic silencing. Genome Biol. 12, 133 (2011). 24. Dawson, M.A. et al. Inhibition of BET recruitment to chromatin as an effective treatment for MLL-fusion leukaemia. Nature 478, 529–533 (2011). 25. Haber, D.A., Gray, N.S. & Baselga, J. The evolving war on cancer. Cell 145, 19–24 (2011). 26. Sharma, S.V. et al. A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell 141, 69–80 (2010).
370
VOLUME 46 | NUMBER 4 | APRIL 2014 Nature Genetics
ONLINE METHODS
npg
© 2014 Nature America, Inc. All rights reserved.
Cell culture. The human T cell leukemia cell lines DND-41 and KOPT-K1 were a kind gift of A.T. Look (Dana-Farber Cancer Institute). Cells were grown in RPMI1640 containing 10% FCS at 37 °C with 5% CO2 , tested for mycoplasma at regular intervals and maintained at a density of between 5 × 105 and 2 × 106 cells/ml. Persister cells were established by treating DND-41 or KOPT-K1 cells with 1 µM GSI for at least 7 weeks, replenishing the inhibitor every 3–4 d (Compound E, EMD4 Biosciences). Reversed cells were generated from persister cells by culturing without GSI for a minimum of 2 weeks. Cell numbers were determined using the Nexcelom Bioscience Cellometer Auto T4. Single cells were sorted into 96-well round-bottom plates and continuously treated with 1 µM Compound E or control. The number of clones that grew out after 4–6 weeks of culture was determined using a Zeiss Observer.Z1 with AxioCam MRm with a 2.5× objective imaging system. At least 500 wells were analyzed per condition in 2 independent experiments. Primary T-ALL cells obtained from pediatric patients with T-ALL with institutional review board (IRB) approval were passaged through NOD-scid IL2Rgnull (NSG) mice and maintained in MEMα containing 10% FCS, 10% heatinactivated human AB+ serum, 10 ng/ml human interleukin (IL)-7, 50 ng/ml human stem cell factor (SCF), 20 ng/ml human FLT3 ligand, 20 nM insulin and 100 ng/ml IL-2 at 37 °C with 5% CO2. Cell proliferation and apoptosis assays. Viable T-ALL cell lines were plated in 2–4 replicates in black or white opaque flat-bottom 96-well tissue culture plates. For proliferation assays, cells were titrated to allow log-phase growth for a period of 4 to 9 d before readout (5,000 to 15,000 cells per well). Viable primary T-ALL cells were plated in 2–3 replicates in black opaque flat-bottom 96-well tissue culture plates at 50,000–100,000 cells per well and were cultured for 6 d. Cell proliferation was measured using the CellTiter-Glo Luminescent Cell Viability Assay from Promega as described by the manufacturer. Apoptosis was measured as a function of caspase 3 and caspase 7 cleavage using the Caspase-Glo 3/7 Luminescent Assay from Promega as described by the manufacturer. End-point luminescence was measured on a SpectraMax M5 plate reader (Molecular Diagnostics). Statistical analyses were performed using Student’s t tests, assuming equal variance. A two-sided P value of
