University of Groningen Evolution of karyotype landscapes in cancer Bakker, Bjorn

(1)

University of Groningen

Evolution of karyotype landscapes in cancer

Bakker, Bjorn

DOI:

10.33612/diss.166886747

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2021

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Bakker, B. (2021). Evolution of karyotype landscapes in cancer. University of Groningen. https://doi.org/10.33612/diss.166886747

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

553979-L-bw-Bakker 553979-L-bw-Bakker 553979-L-bw-Bakker 553979-L-bw-Bakker Processed on: 25-2-2021 Processed on: 25-2-2021 Processed on: 25-2-2021

Processed on: 25-2-2021 PDF page: 1PDF page: 1PDF page: 1PDF page: 1

EVOLUTION OF

KARYOTYPE LANDSCAPES

IN CANCER

(3)

The work described in this thesis was conducted at the European Research Institute for the Biology of Ageing (ERIBA) in the Laboratory of Genomic Instability in Development and Disease, University of Groningen, Groningen, the Netherlands.

ISBN: 978-94-6419-157-8

Cover design: Ineke de Haan & Charles Darwin Lay-out: Gildeprint

Printing: Gildeprint

(4)

Evolution of

karyotype landscapes

in cancer

PhD thesis

to obtain the degree of PhD at the University of Groningen

on the authority of the Rector Magnificus Prof. C. Wijmenga

and in accordance with the decision by the College of Deans. This thesis will be defended in public on

Wednesday 21st_{of April at 14:30 hours}

by

Bjorn Bakker

born on 2 December 1989

(5)

Promotores

Prof. dr. ir. F. Foijer Prof. dr. G. de Haan

Assessment committee

Prof. dr. B.M. Bakker Prof. dr. J.J. Schuringa Prof. dr. L. Wessels

(6)

Paranymphs

Arthur F. Svendsen Daniëlle G. Luinenburg

(7)

TABLE OF CONTENTS

General introduction and thesis outline 8

Chapter 1

Introduction: CINcere modelling: what have mouse models for chromosome

instability taught us? 13

This chapter is based on: Recent Results Cancer Res. 2015;200:39-60

Chapter 2

How to count chromosomes in a cell: an overview of current and novel

technologies 35

This chapter is based on: Bioessays. 2015; 37: 570-577

Chapter 3

Single-cell sequencing reveals karyotype heterogeneity in murine and

human malignancies 51

This chapter is based on: Genome Biology. 2016; 17-115

Chapter 4

Deletion of the MAD2L1 spindle assembly checkpoint gene is tolerated

in mouse models of acute T-cell lymphoma and hepatocellular carcinoma 77

This chapter is based on: ELife. 2017; 6-e200873

Chapter 5

Predicting CIN rates from single-cell whole genome sequencing data using

an in silico model 111

Manuscript in preparation

Chapter 6

Transient genomic instability drives tumorigenesis through accelerated

clonal evolution 141

In submission

Chapter 7

Single cell DNA sequencing reveals distinct molecular types of basal cell

carcinoma with unique transcriptome features 171

(8)

Chapter 8

Addendum: Extensive clonal branching shapes the evolutionary history of

high-risk pediatric cancers 187

This chapter is based on: Cancer Research. 2020;80:1512–23

General discussion 223 Appendices List of abbreviations 232 Curriculum vitae 233 List of publications 234 Nederlandse samenvatting 238 Acknowledgements/Dankwoord 243

(9)

GENERAL INTRODUCTION AND THESIS OUTLINE

During cell division DNA, the genetic code which is distributed over chromosomes, is duplicated in the S-phase of the cell cycle and equally distributed over two daughter cells in mitosis. In normal tissues chromosome segregation is tightly regulated. However, cancer cells frequently mis-segregate their chromosomes into the daughter cells and this results in aneuploidy, an abnormal number of chromosomes in a cell.

Aneuploidy usually leads to impaired cellular growth, senescence, increased metabolic stress, or even cell death. For this reason, aneuploidy is uncommon in healthy tissues. When cells undergo chromosome mis-segregations at a higher rate than normal this is termed chromosomal instability (CIN). Surprisingly, despite their association with detrimental cell growth, both aneuploidy and CIN are commonly observed in proliferating tumour cells. It has therefore been suggested that CIN provides adaptive capacity to cells that enables rapid emergence of genomic alterations that enhance tumorigenesis or metastatic potential. It is well-established that aneuploidy and CIN are hallmarks of various cancer types. Over the last years many elegant studies have shown how CIN contributes to intra-tumour heterogeneity, metastasis and therapy resistance. However, a better understanding of the dynamics of karyotype evolution is still needed to better understand how CIN drives tumour progression, as most of the current techniques that try to map tumour evolution focus on the end stage of the tumour karyotypes.

In this thesis I have aimed to better map karyotype evolution in tumour cells that undergo continuous CIN using advanced single cell sequencing tools and state-of-the-art cell and mouse models.

Over the past 20 years or so, various tools and model systems have been engineered to study CIN in vivo. Chapter 1 discusses mouse models that have been engineered to study CIN in vivo. Ongoing chromosome mis-segregations are typically provoked by mutating or knocking out components of the cellular machinery involved in faithful chromosome segregation during mitosis. The majority of the mouse models described in this chapter show that full inactivation of genes that ensure faithful chromosome segregation (including spindle assembly checkpoint (SAC) proteins) lead to early embryonic lethality. To circumvent this, conditional mouse models have been developed by our lab (described in this thesis) and others to study the effects of CIN and the resulting aneuploidy. Most

(10)

models suggest that CIN alone is not sufficient to induce tumorigenesis, but that additional, cancer-predisposing hits are needed. To address why this is the case, appropriate techniques for karyotyping are required that are discussed in the next chapter.

Chapter 2 contrasts and compares the strengths and limitations of both existing and novel

cytogenetic techniques required to address questions outlined in Chapter 1. The ideal method enables researchers to determine 1) the complete karyotype of 2) individual cells to assess heterogeneity 3) without a bias towards dividing cells. Conventional karyotyping techniques typically only meet two out of three of these criteria, as they are either limited to a few chromosomes per cell, require bulk material and thus does not include single-cell resolution, or can only be applied to metaphase spreads and thus require dividing cells. Single-cell DNA sequencing (scWGS) is proposed to be costly, but one of the most suitable methods to examine intra-tumour karyotype heterogeneity as it meets all three criteria. I therefore, in my thesis, developed tools for scWGS to study karyotype heterogeneity as it is a suitable technique to address my research questions. The focus of the following chapters is the development and application of (tools for) scWGS.

In chapter 3 I describe the single-cell genome sequencing platform developed at ERIBA and data analysis tool AneuFinder that I co-developed. I use this pipeline in the subsequent chapters of my thesis, and many more collaborative studies not in my thesis, to detect copy number alterations in tumour samples. Using this pipeline we examined the karyotypes of murine T-ALLs driven by mutation of the spindle assembly checkpoint protein Mps1. Previous analysis of the model revealed an attenuated spindle assembly checkpoint resulted in ongoing CIN. However, the bulk karyotypes as determined by aCGH suggested the tumour cells had acquired a stable clonal karyotype with recurrent chromosome aberrations. By applying the scWGS platform to these T-ALLs we found that they display a CIN phenotype, evidenced by large-scale cell-to-cell variability in karyotypes. Furthermore, when pressured, tumours select for new karyotypes to adjust to the new conditions, another indication of ongoing CIN. Importantly, the development of the scWGS platform at ERIBA has been important for the CIN field and has supported multiple studies from other labs to gain insight in aneuploidy and CIN on a single cell level in various tumour models (also see addendum/publication list).

We next explored the effects on karyotype dynamics in the background of a different CIN driver. Chapter 4 describes a mouse model for CIN-driven T-ALL and hepatocellular carcinoma (HCC) as a result of loss of the spindle assembly checkpoint protein Mad2.

(11)

Similar to what is described in chapter 3, we find that inactivation of Mad2 provokes ongoing CIN and accelerates the onset of tumorigenesis in a p53-null background in both T-cells and liver. Although the driver of CIN was the same in both tissues, we found evidence of tissue-specific selection for karyotype aberrations. Comparative genomics suggests that the tumour cells select for aberrations of chromosomes that carry oncogenes or tumour suppressor genes.

To better understand the karyotype dynamics of cancer cells undergoing continuous CIN I developed a forward-stochastic model for karyotype evolution that is described in

chapter 5. Using this simulation model, we can accurately predict the CIN rate in a set

of mouse T-ALLs using only data on chromosome copy number frequency. The model suggests that the favoured tumour karyotypes emerge early during tumour development. In contrast to what has been reported previously, we found that whole genome duplication is not always an efficient step for karyotype evolution in cancer, but only in some cases. We expect that our tool will be used in the near future to predict CIN in primary patient data and thereby improve cancer risk stratification.

To further explore karyotype evolution during early tumour development, in chapter 6 we compared CIN in two mouse models (conditional Plk4 overexpression vs. conditional Mad2 inactivation, both in a p53-null background) in collaboration with the lab of Don Cleveland. Both transient (through Plk4 overexpression) as well as continuous CIN (through Mad2-loss) drive early development and selection for cancer cells with recurrent favourable karyotype aberrations. Interestingly, transient CIN tumours show less intratumor heterogeneity, but also acquire the optimal karyotype earlier in their development. In both models these recurrent karyotypes are thought to enhance expression of important T-ALL driver genes, including Myc on chromosome 15. Together this data reinforces the idea that T-ALL cells evolve towards a karyotype landscape that offers the most optimal fitness. To investigate the role of CIN and aneuploidy in human primary cancers we have focused in chapter 7 on basal cell carcinoma (BCC), a tumour of the skin. In this study we find that BCCs can be stratified based on karyotypic makeup. Traditionally, BCCs were classified solely on their histopathological features. We propose that scWGS can provide insight in the karyotype landscape of human BCC samples and can support in BCC classification, which is supported by distinct transcriptional features in the scWGS-stratified BCCs. Altogether, the data in this chapter suggests that cancers driven by CIN are different from more euploid tumours, which could affect their response to treatment.

(12)

In chapter 8, as an addendum I highlight an application of single-cell karyotyping for patient risk stratification of three types of paediatric cancer: neuroblastoma, rhabdomyosarcoma, and Wilms tumour. In collaboration with David Gisselsson we established that the risk of recurrence is proportional to the degree of genomic heterogeneity (both mutational and karyotypic). In addition, we found that mutations arising early in neuroblastoma and rhabdomyosarcoma affect the risk of relapse, whereas in Wilms tumours such events occur later in tumour development.

(13)