LETTER
https://doi.org/10.1038/s41586-019-1472-0
BORIS promotes chromatin regulatory interactions
in treatment-resistant cancer cells
David N. Debruyne1,2,15, Ruben Dries1,2,3,15, Satyaki Sengupta1,2, Davide Seruggia1,4,5,6, Yang Gao1,2, Bandana Sharma1,2,
Hao Huang1,2, Lisa Moreau^7 , Michael McLane1,2, Daniel S. Day8,9, Eugenio Marco3,10, Ting Chen^11 , Nathanael S. Gray12,13,
Kw o k - K i n Wo n g^14 , Stuart H. Orkin1,4,5,6, Guo-Cheng Yuan3,10, Richard A. Young8,9 & Rani E. George1,2,4*
The CCCTC-binding factor (CTCF), which anchors DNA loops that
organize the genome into structural domains, has a central role in
gene control by facilitating or constraining interactions between
genes and their regulatory elements^1 ,^2. In cancer cells, the disruption
of CTCF binding at specific loci by somatic mutation^3 ,^4 or DNA
hypermethylation^5 results in the loss of loop anchors and consequent
activation of oncogenes. By contrast, the germ-cell-specific
paralogue of CTCF, BORIS (brother of the regulator of imprinted
sites, also known as CTCFL)^6 , is overexpressed in several cancers^7 –^9 ,
but its contributions to the malignant phenotype remain unclear.
Here we show that aberrant upregulation of BORIS promotes
chromatin interactions in ALK-mutated, MYCN-amplified
neuroblastoma^10 cells that develop resistance to ALK inhibition.
These cells are reprogrammed to a distinct phenotypic state
during the acquisition of resistance, a process defined by the initial
loss of MYCN expression followed by subsequent overexpression
of BORIS and a concomitant switch in cellular dependence from
MYCN to BORIS. The resultant BORIS-regulated alterations in
chromatin looping lead to the formation of super-enhancers that
drive the ectopic expression of a subset of proneural transcription
factors that ultimately define the resistance phenotype. These results
identify a previously unrecognized role of BORIS—to promote
regulatory chromatin interactions that support specific cancer
phenotypes.(^1) Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, MA, USA. (^2) Department of Pediatrics, Harvard Medical School, Boston, MA, USA. (^3) Department of Biostatistics and
Computational Biology, Dana-Farber Cancer Institute, Boston, MA, USA.^4 Division of Hematology/Oncology, Boston Children’s Hospital, Boston, MA, USA.^5 Harvard Stem Cell Institute, Harvard
Medical School, Boston, MA, USA.^6 Howard Hughes Medical Institute, Boston, MA, USA.^7 Department of Radiation Oncology, Dana-Farber Cancer Institute, Boston, MA, USA.^8 Whitehead Institute
for Biomedical Research, Cambridge, MA, USA.^9 MIT Department of Biology, Cambridge, MA, USA.^10 Department of Biostatistics, Harvard TC Chan School of Public Health, Boston, MA, USA.
(^11) Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA. (^12) Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA, USA. (^13) Department of Biological
Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA, USA.^14 Division of Hematology and Medical Oncology, Laura and Isaac Perlmutter Cancer Center, New York
University Langone Medical Center, New York, NY, USA.^15 These authors contributed equally: David N. Debruyne, Ruben Dries.. *e-mail: [email protected]
abc
d
f
g
e
0
0
0.02
0.01
–0.01
–0.02
PC1
PC2
0
2.5
5.0
7.5
10.0
Rescaled expression
BORIS
SOX9
SOX2
MYCN
Pseudotime
Sens 1Sens 2IR 1IR 2IR 3Res 1Res 2Res 3Res 4
BORIS
MYCN
Actin
MYCNKD/BORISInd
Sensitive
DMSO
TAE
DOX + DMSO
DOX + TAE
Sensitive
Resistant
Resistant + shBORIS
Resistant + shCtrl
PC1
PC2
–50 –25 0525 0
–30
0
30
40 nM (IC 50 )
0 (months) 12
Sensitive
–2 0 +2
z-score
Resistant
PKIB
NEUROD4
PLP1
CER1
PTX3
INHBE
MYL1
SIX1
RPRM
BORIS
KCTD8
CEBPD
ARHGAP36
DACT1
RBM24
NEUROD1
NEUROG1
INSM1
KLHL1
SNAI2
chr1
RPM per bp
4k
2k
0
chr2chr3chr4chr5chr6chr7chr8chr9chr10chr11chr12chr13chr14chr15chr16chr17chr18chr19chr20chr21chr22chrXchrY
10 –3 10 –2 10 –1 100 101
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
log([TAE684] (μM))
Relative viability
Sensitive (39 nM)
Resistant (1,210 nM)
10 –3 10 –2 10 –1 100 101
0
0.2
0.4
0.6
0.8
1.0
1.2
log([TAE684] (μM))
Relative viability
TAE (24.4 nM)
DOX + DMSO (74.1 nM)
DOX + TAE (525.8 nM)
DMSO (35.3 nM)
Resistant
Resistant
IR
Sensitive
[TAE684] Sensitive
–0.01 0.01
1,200 nM
Sensitive Resistant
Fig. 1 | Targeted therapy resistance in
neuroblastoma is associated with
transcriptional reprogramming and a switch
in dependency from amplified MYCN to
BORIS. a, Top, schematic representation of the
development of resistance. Bottom, dose–response
curves of TAE684-sensitive and -resistant Kelly
neuroblastoma cells incubated in increasing
concentrations of TAE684 for 72 h. Data are
mean ± s.d., n = 3 biological replicates. b, Heat
map of gene expression values in sensitive versus
resistant cells (n = 2 biological replicates). Rows
are z-scores calculated for each gene in both cell
types. c, PCA of scRNA-seq data of sensitive
(n = 5,432), intermediate resistant (IR; n = 6,376)
and resistant (n = 6,379) cells showing the first
two principal components (PCs). d, Pseudotime
analysis of transcription factor expression during
the development of resistance. e, ChIP–seq signals
of genome-wide MYCN binding in sensitive
and resistant cells, reported as reads per million
(RPM) per base pair (bp) for each chromosome
(chr). f, PCA of gene expression profiles showing
the first two principal components (n = 2
biological replicates). g, Dose–response curves for
TAE684 (half-maximum inhibitory concentration
(IC 50 ) values in parenthesis) and immunoblot
analysis (representative of two independent
experiments) of BORIS and MYCN expression
in sensitive cells expressing short hairpin
RNA (shRNA) against MYCN (MYCNKD) and
doxycycline-inducible BORIS (BORISInd), treated
with dimethylsulfoxide (DMSO) or 1 μM TAE684,
with or without doxycycline (DOX). Data are
mean ± s.d., n = 3 biological replicates.
676 | NAT URE | VOL 572 | 29 AUGUST 2019