. 2018 Dec 10;34(6):939-953.e9.

doi: 10.1016/j.ccell.2018.10.014. Epub 2018 Nov 21.

KDM5 Histone Demethylase Activity Links Cellular Transcriptomic Heterogeneity to Therapeutic Resistance

Kunihiko Hinohara¹, Hua-Jun Wu², Sébastien Vigneau³, Thomas O McDonald⁴, Kyomi J Igarashi³, Kimiyo N Yamamoto², Thomas Madsen², Anne Fassl³, Shawn B Egri⁵, Malvina Papanastasiou⁵, Lina Ding¹, Guillermo Peluffo¹, Ofir Cohen⁶, Stephen C Kales⁷, Madhu Lal-Nag⁷, Ganesha Rai⁷, David J Maloney⁷, Ajit Jadhav⁷, Anton Simeonov⁷, Nikhil Wagle⁸, Myles Brown⁹, Alexander Meissner¹⁰, Piotr Sicinski³, Jacob D Jaffe⁵, Rinath Jeselsohn¹, Alexander A Gimelbrant³, Franziska Michor¹¹, Kornelia Polyak¹²

Affiliations

¹ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA.
² Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA 02115, USA; Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA.
³ Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Genetics, Harvard Medical School, Boston, MA 02115, USA.
⁴ Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA 02115, USA; Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA; Center for Cancer Evolution, Dana-Farber Cancer Institute, Boston, MA 02215, USA.
⁵ The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
⁶ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
⁷ National Center for Advancing Translational Sciences, Bethesda, MD 20892, USA.
⁸ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
⁹ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA; Center for Functional Cancer Epigenetics, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Ludwig Center at Harvard, Boston, MA 02215, USA.
¹⁰ Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
¹¹ Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA 02115, USA; Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA; Center for Cancer Evolution, Dana-Farber Cancer Institute, Boston, MA 02215, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA; Ludwig Center at Harvard, Boston, MA 02215, USA. Electronic address: michor@jimmy.harvard.edu.
¹² Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA; Center for Cancer Evolution, Dana-Farber Cancer Institute, Boston, MA 02215, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA; Center for Functional Cancer Epigenetics, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Ludwig Center at Harvard, Boston, MA 02215, USA. Electronic address: kornelia_polyak@dfci.harvard.edu.

PMID: 30472020
PMCID: PMC6310147
DOI: 10.1016/j.ccell.2018.10.014

KDM5 Histone Demethylase Activity Links Cellular Transcriptomic Heterogeneity to Therapeutic Resistance

Kunihiko Hinohara et al. Cancer Cell. 2018.

. 2018 Dec 10;34(6):939-953.e9.

doi: 10.1016/j.ccell.2018.10.014. Epub 2018 Nov 21.

Authors

Affiliations

¹ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA.
² Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA 02115, USA; Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA.
³ Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Genetics, Harvard Medical School, Boston, MA 02115, USA.
⁴ Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA 02115, USA; Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA; Center for Cancer Evolution, Dana-Farber Cancer Institute, Boston, MA 02215, USA.
⁵ The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
⁶ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
⁷ National Center for Advancing Translational Sciences, Bethesda, MD 20892, USA.
⁸ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
⁹ Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA; Center for Functional Cancer Epigenetics, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Ludwig Center at Harvard, Boston, MA 02215, USA.
¹⁰ Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA.
¹¹ Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA 02115, USA; Department of Stem Cell and Regenerative Biology, Harvard University, Cambridge, MA 02138, USA; Center for Cancer Evolution, Dana-Farber Cancer Institute, Boston, MA 02215, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA; Ludwig Center at Harvard, Boston, MA 02215, USA. Electronic address: michor@jimmy.harvard.edu.
¹² Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Department of Medicine, Harvard Medical School, Boston, MA 02115, USA; Center for Cancer Evolution, Dana-Farber Cancer Institute, Boston, MA 02215, USA; The Eli and Edythe L Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA; Center for Functional Cancer Epigenetics, Dana-Farber Cancer Institute, Boston, MA 02215, USA; Ludwig Center at Harvard, Boston, MA 02215, USA. Electronic address: kornelia_polyak@dfci.harvard.edu.

PMID: 30472020
PMCID: PMC6310147
DOI: 10.1016/j.ccell.2018.10.014

Erratum in

KDM5 Histone Demethylase Activity Links Cellular Transcriptomic Heterogeneity to Therapeutic Resistance.
Hinohara K, Wu HJ, Sébastien Vigneau, McDonald TO, Igarashi KJ, Yamamoto KN, Madsen T, Fassl A, Egri SB, Papanastasiou M, Ding L, Peluffo G, Cohen O, Kales SC, Lal-Nag M, Rai G, Maloney DJ, Jadhav A, Simeonov A, Wagle N, Brown M, Meissner A, Sicinski P, Jaffe JD, Jeselsohn R, Gimelbrant AA, Michor F, Polyak K. Hinohara K, et al. Cancer Cell. 2019 Feb 11;35(2):330-332. doi: 10.1016/j.ccell.2019.01.012. Cancer Cell. 2019. PMID: 30753830 Free PMC article. No abstract available.

Abstract

Members of the KDM5 histone H3 lysine 4 demethylase family are associated with therapeutic resistance, including endocrine resistance in breast cancer, but the underlying mechanism is poorly defined. Here we show that genetic deletion of KDM5A/B or inhibition of KDM5 activity increases sensitivity to anti-estrogens by modulating estrogen receptor (ER) signaling and by decreasing cellular transcriptomic heterogeneity. Higher KDM5B expression levels are associated with higher transcriptomic heterogeneity and poor prognosis in ER⁺ breast tumors. Single-cell RNA sequencing, cellular barcoding, and mathematical modeling demonstrate that endocrine resistance is due to selection for pre-existing genetically distinct cells, while KDM5 inhibitor resistance is acquired. Our findings highlight the importance of cellular phenotypic heterogeneity in therapeutic resistance and identify KDM5A/B as key regulators of this process.

Keywords: KDM5B; acquired resistance; barcoding; cellular heterogeneity; endocrine resistance; epigenetic; pre-existing resistance; single-cell RNA-seq; subclonal fraction; transcriptomic heterogeneity.

PubMed Disclaimer

Conflict of interest statement

COMPETING FINANCIAL INTERESTS

Figures

**Figure 1.. The role of KDM5B and KDM5A on endocrine therapies and transcriptomic variability.**
(A) Cellular viability after fulvestrant treatment of parental MCF7, *KDM5B*-KO and *KDM5A*-KO cells. (B) Cellular viability after fulvestrant treatment of a panel of breast cancer cell lines pre-treated with DMSO or KDM5i. (C) Graph depicting percent change in tumor volume from baseline in control, fulvestrant (FULV), C48, and combined treatment groups. Black line marks 30% decrease in volume, which is commonly used as a cut off to define response in clinical studies. (D) Representative MRI images of tumors before and after treatment in vehicle and combined C48+FULV group. (E) Representative immunofluorescence analysis of the indicated markers in tumors of the four treatment groups. Scale bar, 100 µm. (F) Graphs depicting quantification of immunofluorescence images. In (A) and (B), Error bars represent s.d., n = 6. See also Figure S1 and Table S1.

**Figure 2.. KDM5 activity and H3K4me3 peak broadness.**
(A) H3K4me3 and H3K4me2 peak width plotted against peak height before and at different time points (day 0–14) after treatment with KDM5-C70 inhibitor. Mean values are shown as dotted lines. Shaded areas indicate interquartile range. (B) Gene tracks depicting KDM5B and H3K4me3 signal at selected genomic loci. X-axis shows position along the chromosome with gene structures drawn below, whereas y-axis shows genomic occupancy in units of reads per million reads (RPM). (C) Correlation between promoter H3K4me3 peak broadness changes and changes in percent of cells expressing the corresponding gene in KDM5C70-treated cells. Enrichment analysis of H3K4me3 width increase in C70 is performed against the genes with increased percent of expressing cells in C70 for all genes or genes without expression change. H3K4me3 width changes are calculated as the average width changes across all six cell lines. *** fdr<0.001; ** fdr<0.01; * fdr<0.25. (D) Plot depicting percentage of cells expressing *ZMYND8* in MCF7 and C70-treated MCF7 cells. All single cells are ranked and grouped into 10 groups based on their sequence depth to avoid variability due to this. The percent of expressing cells is calculated for each group, and a weighted t-test is performed to access the significance of the difference between two samples. The box indicates the interquartile range (IQR), the line inside the box shows the median and whiskers show the locations of either 1.5×IQR above the third quartile or 1.5×IQR below the first quartile. See also Figure S2.

**Figure 3.. KDM5 activity and transcriptomic heterogeneity.**
(A) Gini index of single-cell inDrop data. The distribution of Gini coefficients of all genes in each sample is shown as grey density plot. Selected luminal (blue), basal/mesenchymal (red), KDM5i-induced (green), and housekeeping (black) genes are highlighted. (B) Violin plot showing distribution of normalized expression of *ZMYND8* based on single cell RNA-seq data. Dots within violin represent the transcript counts in single cells. The “-” and “+” inside the violin indicate the median and mean values, respectively. (C) Graphs depicting cell-to-cell distance in the indicated cell populations. Wilcoxon rank-sum test p values are shown. Note the analysis of pair-wised distances between all single cells generates a large number of data points, which makes the p value less informative than the relative differences between mean values (shown on the right side) and box profiles. The box indicates the interquartile range (IQR), the line inside the box shows the median and whiskers show the locations of either 1.5×IQR above the third quartile or 1.5×IQR below the first quartile. (D) Plot depicting the number of genes with changes in Gini index after KDM5-C70 treatment. (E) Top signaling pathways enriched among genes with decreasing Gini index after KDM5-C70 treatment in MCF7 and FULVR cells. (F) Shannon’s equitability showing a correlation between *KDM5B* gene expression and transcriptomic heterogeneity in ER⁺ (n = 808) breast tumors in the TCGA data set. All tumors are stratified into four groups with identical sample size based on *KDM5B* expression levels from low (1) to high (4). (G) Shannon’s equitability showing a correlation between *KDM5B* gene expression and transcriptomic heterogeneity in ER⁺ (n = 108) distant metastases of breast cancer in the Metastatic Breast Cancer Project data set. Patient stratification is as same as in (F). (H) Patient survival between high and low transcriptome heterogeneity in all (n = 1,093), ER⁺ (n = 808) and ER⁻ (n = 237) breast tumors in the TCGA data. All patients are stratified into two groups with identical sample size based on the transcriptome heterogeneity. In (F) and (G), the outer violin indicates the entire distribution, the inner violin in white indicates the interquartile range, the “.” and “+” inside the violin show the median and mean value, respectively. See also Figure S3.

**Figure 4.. Characterization of acquired KDM5i resistance.**
(A) Cellular viability of MCF7, C70R and, C49R cells after treatment with C70 or C49. (B) Morphology of MCF7, C70R, and C49R cells. Scale bars, 100 µm. (C) GSEA plots depicting the relationship between genes in C70R cells and genes in endocrine resistant cells. Genes are ranked by the statistical significance of differential expression analysis between MCF7 and endocrine resistant cells (FULVR and TAMR) in x-axis, with up genes in endocrine resistant cells on the left side. The enrichment score of top 500 up or down genes in C70R compared to MCF7 cells are plotted as red and blue curve, respectively. (D) Cellular viability after treatment with C70 or C49 in FULVR, TAMR, and in MCF7-*ESR1*^Y537S cells. (E) Colony growth of MCF7 and KDM5i-resistant cells in charcoal-stripped medium. (F) Immunoblot for the indicated proteins following E2 treatment. (G) ER chromatin binding peaks (±500 bp peak summit) in MCF7, C49R, and C70R cells after estrogen deprivation (0 min) and 45 min after E2 treatment. Only the ER binding peaks responding to E2 treatment in MCF7 cells are shown. (H) Integrated analysis of associations between gene expression changes at different time points (0–6 hr) after E2 treatment and ER chromatin binding in the indicated clusters and cell lines. The box indicates the interquartile range (IQR), the line inside the box shows the median and whiskers show the locations of either 1.5×IQR above the third quartile or 1.5×IQR below the first quartile. In (A) and (D), Error bars represent s.d., n = 6. See also Figure S4 and Tables S2, S3, S4.

**Figure 5.. Single cell profiling of drug-resistant cells.**
(A) Hexagonal plots depicting the bootstrap classification of single cells in populations of MCF7, fulvestrant-treated (MCF7+FULV), and FULVR cells. Each point is one single cell and is positioned along axes according to its bootstrapping classification score for the indicated cell identity. Black, green, and blue cells are classified as MCF7, MCF7+FULV, and FULVR cells, and grey cells are unclassified. A few cells are classified as combination of two cell identities and are represented by mixed color of the two, and positioned at the edges of 2, 6, and 10 o’clock. (B) Hexagonal plots depicting the bootstrap classification of single cells in populations of MCF7, C70-treated MCF7 (MCF7+C70), and C70R cells. Each point is one single cell and is positioned along axes according to its bootstrapping classification score for the indicated cell identity. Black, light blue, and red cells are classified as MCF7, MCF7+C70, and C70R cells, and grey cells are unclassified. A few cells are classified as combination of two cell identities and are represented by mixed color of the two, and positioned at the edges of 2, 6, and 10 o’clock. (C) Projection of SPADE tree for each cell line. Colors and size of the node correspond to the percentage of cells that belongs to a given cluster. Light gray dots mark cells with low marker expression in all channels. (D) Relative proportions of cells in FULVR population with MCF7, MCF7+C70, and C70R gene signature. (E) Relative proportions of cells in C70R population with MCF7, MCF7+FULV, and FULVR gene signature. See also Figure S5 and Table S5.

**Figure 6.. Resistance to anti-estrogens and KDM5i in MCF7 cells.**
(A) Cellular viability after treatment with C70 and C49, fulvestrant or tamoxifen in parental and cells with acquired resistance to the indicated agents. Error bars represent s.d., n = 6. (B) Bar graph depicting percentage of unique barcodes in FULVR and TAMR relative to parental MCF7 cells at same passage. (C) Pie chart depicting percentage of barcodes overlapping between MCF7 and FULVR/TAMR cells. (D) Bar graph depicting percentage of total barcodes shared among all replicates in each of the indicated cell populations. (E) Pie chart depicting percentage of barcodes overlapping between FULVR and TAMR. (F) Bar graph depicting percentage of unique barcodes in C70R and C49R relative to MCF7 cells at same passage. (G) Pie chart depicting percentage of barcodes overlapping between MCF7 and C70R/C49R cells. (H) Panels show model-predicted percentages of total barcodes shared by quadruplicates after simulation for different mutation probabilities (µ) and seeded fractions of preexisting resistant barcodes (ρ) in the treatment with the indicated inhibitors compared to the same statistic from the experimental data (horizontal line). The growth rates in simulations were based on experimental data. (I) Mutated genes detected in resistant but not in MCF7 cells. Colors and stars indicate the type of mutations and significance of downstream GSEA in the corresponding resistant cell lines, respectively. The significance of downstream GSEA represents the downstream genes of mutations are significantly enriched in up/down regulated genes in the corresponding resistant cell lines. See also Figure S6 and Table S6.

See this image and copyright information in PMC

References

1. Anders S, Pyl PT, and Huber W (2015). HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169. - PMC - PubMed
1. Ben-David U, Siranosian B, Ha G, Tang H, Oren Y, Hinohara K, Strathdee CA, Dempster J, Lyons NJ, Burns R, et al. (2018). Genetic and transcriptional evolution alters cancer cell line drug response. Nature 560, 325–330. - PMC - PubMed
1. Benayoun BA, Pollina EA, Ucar D, Mahmoudi S, Karra K, Wong ED, Devarajan K, Daugherty AC, Kundaje AB, Mancini E, et al. (2014). H3K4me3 breadth is linked to cell identity and transcriptional consistency. Cell 158, 673–688. - PMC - PubMed
1. Bendall SC, Simonds EF, Qiu P, Amir el AD, Krutzik PO, Finck R, Bruggner RV, Melamed R, Trejo A, Ornatsky OI, et al. (2011). Single-cell mass cytometry of differential immune and drug responses across a human hematopoietic continuum. Science 332, 687–696. - PMC - PubMed
1. Bhang HE, Ruddy DA, Krishnamurthy Radhakrishna V, Caushi JX, Zhao R, Hims MM, Singh AP, Kao I, Rakiec D, Shaw P, et al. (2015). Studying clonal dynamics in response to cancer therapy using high-complexity barcoding. Nat Med 21, 440–448. - PubMed

Publication types

Actions
Actions

MeSH terms

Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions

Substances

Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

KDM5 Histone Demethylase Activity Links Cellular Transcriptomic Heterogeneity to Therapeutic Resistance

Affiliations

KDM5 Histone Demethylase Activity Links Cellular Transcriptomic Heterogeneity to Therapeutic Resistance

Authors

Affiliations

Erratum in

Abstract

Conflict of interest statement

Figures

References

Publication types

MeSH terms

Substances

Grants and funding