University of Southern California Dissertations and Theses

Molecular signatures underlying intercellular differences in leukemia progression and chemotherapy response

(USC Thesis Other)

Molecular signatures underlying intercellular differences in leukemia progression and chemotherapy response

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Molecular signatures underlying intercellular differences in
leukemia progression and chemotherapy response

by
HUMBERTO CONTRERAS-TRUJILLO

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
Development, Stem Cells and Regenerative Medicine

May 2020
ii

1 PREFACE
Abstract
Cellular heterogeneity is a cause of treatment resistance in cancer; however, its underlying
mechanisms are largely unknown. Here, we combined in vivo single-cell tracking with single-cell
mRNA sequencing to study cellular heterogeneity in a patient-derived xenograft (PDX) model of
B-cell acute lymphoblastic leukemia (B-ALL). Comparing PDX mice from serial transplantation,
we demonstrated that variability in clonal expansion was largely cell autonomous and associated
with differential gene expression. Clonal comparison between tissues revealed marked anatomical
biases surprising for a “liquid” tumor, which has important implications for clinical diagnostic
sampling. Furthermore, small subsets of B-ALL clones with distinct gene expression signatures
responded differently to intensive and maintenance chemotherapies. Integrated analyses of
intercellular differences on cellular and molecular activities revealed new insights into cancer
progression.

iii
Acknowledgments
I thank my thesis advisor Dr. Rong Lu, for her mentorship throughout my graduate studies.
Rong, I am thankful for all your support and advice. The past five years have been a great learning
experience. With your mentorship, I have developed important professional skills from scientific
writing to presenting. Thank you! I also thank the rest of my doctoral committee members Dr. Qi-
Long Ying and Dr. Min Yu as well as Dr. Ruchi Bajpai and Dr. Akil Merchant for their guidance
and advice. All of their support has been essential to my training at USC.
I am also extremely grateful to Dr. Irving Weissman, Dr. Jens-Peter Volkmer and Dr.
Stephen Willingham, whom I worked with during a California Institute of Regenerative Medicine
fellowship at Stanford School of Medicine. I learned many important techniques in cancer research
under their mentorship that prepared me for my graduate work. I was also inspired to pursue a
career in cancer immunotherapy. Without their mentorship I would not be where I am today.
I thank the rest of the Lu lab, especially Jiya, Samir, Du, Mary and Areen. Thank you for
all of your help. Jiya and Samir, thank you for all of your support with data analysis. Your help
has been instrumental throughout my dissertation. Du, I am very thankful for your contributions
with the single-cell RNA sequencing experiments. Your contributions have been significant in
linking our clonal barcode tracking data with single-cell gene expression. I am also very grateful
to my funding source the Ruth L. Kirschstein National Research Service Award of the National
Cancer Institute (F31CA206463).
I thank my family and friends for their constant support and encouragement. Yigit, thank
you for all of your advice, support and patience.
iv
Dedication Statement

This dissertation is dedicated to:

My mother, Martha Contreras.
“Mamá, gracias por todo tu apoyo, tu amor y por siempre motivarme a salir adelante.”
(Mom, thank you for all of your support, your love, and for always motivating me to excel.)

To the memory of my father, Manuel Contreras.
“Papá, gracias por motivarme desde chico a perseguir una educación. Tus palabras de
sabiduría resuenan conmigo.”
(Dad, thank you for encouraging me early on to pursue a higher education. Your words of
wisdom resonate with me.)

v
2 TABLE OF CONTENTS
PREFACE ...................................................................................................................................... ii
ABSTRACT ....................................................................................................................................... ii
ACKNOWLEDGMENTS...................................................................................................................... iii
DEDICATION STATEMENT ............................................................................................................... iv
TABLE OF CONTENTS ............................................................................................................. v
LIST OF FIGURES ................................................................................................................... viii
LIST OF SUPPLEMENTARY FIGURES ................................................................................. x
LIST OF TABLES ...................................................................................................................... xii
CHAPTER 1 .................................................................................................................................. 1
1.1 ABSTRACT ............................................................................................................................ 1
1.2 INTRODUCTION ..................................................................................................................... 2
Advances in cancer clonal tracking .................................................................................... 2
Genetic barcodes in cancer clonal tracking ....................................................................... 3
Genetic barcode experimental system ................................................................................ 4
1.3 RESULTS ............................................................................................................................... 5
Engraftment of patient derived leukemia ............................................................................ 5
Cell-dose optimization ........................................................................................................ 5
Establishing a barcoded patient derived leukemia xenograft ............................................. 6
In vitro growth phenotypes of B-ALL cells derived from different disease stages ............. 7
In vivo clonal dynamics of B-ALL cells derived from different disease stages .................. 7
Low abundance clones are eliminated over time ................................................................ 8
1.4 DISCUSSION .......................................................................................................................... 9
1.5 FIGURES .............................................................................................................................. 11
1.6 SUPPLEMENTARY FIGURES .................................................................................................. 20
CHAPTER 2 ................................................................................................................................ 23
2.1 ABSTRACT .......................................................................................................................... 23
2.2 INTRODUCTION ................................................................................................................... 24
Cancer clonal evolution .................................................................................................... 24
Patient derived xenograft model ....................................................................................... 24
Cancer clonality through serial transplantation .............................................................. 25

vi
2.3 RESULTS ............................................................................................................................. 26
Clonal selection occurs during serial transplantation ..................................................... 26
Leukemia clonal behaviors during serial transplantation is associated with
gene expression differences .............................................................................................. 26
2.4 DISCUSSION ........................................................................................................................ 27
2.5 FIGURES .............................................................................................................................. 29
CHAPTER 3 ................................................................................................................................ 40
3.1 ABSTRACT .......................................................................................................................... 40
3.2 INTRODUCTION ................................................................................................................... 41
Acute lymphoblastic leukemia........................................................................................... 41
Acute lymphoblastic leukemia extramedullary expansion ................................................ 41
Cancer metastasis ............................................................................................................. 42
CXCL12/CXCR4 ............................................................................................................... 42
3.3 RESULTS ............................................................................................................................. 43
High leukemia clonal correlation between blood, spleen and bone marrow ................... 43
Spatially confined expansion of leukemia clones in the bone marrow ............................. 44
Leukemia clonal expansion in the bone marrow is associated with differentially
expressed genes ................................................................................................................. 44
Extramedullary expansion of leukemia clones associated with differentially
expressed genes ................................................................................................................. 45
3.4 DISCUSSION ........................................................................................................................ 46
3.5 FIGURES .............................................................................................................................. 48
3.6 SUPPLEMENTARY FIGURES .................................................................................................. 58
CHAPTER 4 ................................................................................................................................ 64
4.1 ABSTRACT .......................................................................................................................... 64
4.2 INTRODUCTION ................................................................................................................... 65
Acute lymphoblastic leukemia........................................................................................... 65
Diagnosis and immunophenotype ..................................................................................... 65
Treatment overview of acute lymphoblastic leukemia ...................................................... 66
Induction and consolidation phase ................................................................................... 67
Maintenance phase ........................................................................................................... 67
Relapsed acute lymphoblastic leukemia ........................................................................... 68
4.3 RESULTS ............................................................................................................................. 69
Experimental model recapitulates clinical ALL therapy in PDX ..................................... 69
Clonal emergence under chemotherapy treatment ........................................................... 70
Some B-ALL clones responded differentially to intensive and maintenance
chemotherapy .................................................................................................................... 71
Differentially expressed genes associated with clonal response to chemotherapy .......... 71
4.4 DISCUSSION ........................................................................................................................ 73
vii
4.5 FIGURES .............................................................................................................................. 75
4.6 SUPPLEMENTAL FIGURES .................................................................................................... 83
APPENDIX .................................................................................................................................. 96
MATERIALS AND METHODS ........................................................................................................... 96
Human cells ...................................................................................................................... 96
Leukemia cell culture and lentiviral transduction ............................................................ 96
Mice................................................................................................................................... 97
Human B-ALL engraftment ............................................................................................... 97
Bioluminescent imaging .................................................................................................... 98
Blood sample collection and FACS analysis .................................................................... 98
Chemotherapy treatment ................................................................................................... 99
DNA barcode extraction and analysis ............................................................................ 100
Single-cell RNA sequencing and data analysis ............................................................... 100
False positive score calculation...................................................................................... 101
Statistical analysis .......................................................................................................... 101
REFERENCES ................................................................................................................................ 103

viii
3 LIST OF FIGURES
CHAPTER 1
FIGURE 1.1. XENOGRAFT MODEL FOR STUDYING HUMAN LEUKEMIA IN MICE. ............................... 11
FIGURE 1.2. CIRCULATING LEUKEMIA CELLS ARE DETECTED EARLIER WITH HIGHER
CELL DOSE TRANSPLANTATION. .................................................................................. 13
FIGURE 1.3. CLONAL TRACKING OF HUMAN B-ALL CELLS USING GENETIC BARCODES. ................ 14
FIGURE 1.4. LEUKEMIA IN VITRO CULTURES. .................................................................................. 16
FIGURE 1.5. LEUKEMIA CELLS ACQUIRED AT DIFFERENT DISEASE STAGES. .................................... 17
FIGURE 1.6. LEUKEMIA CLONES EXPAND PROPORTIONALLY IN PDX MODEL. ................................ 18
FIGURE 1.7. CLONAL ABUNDANCE CHANGES OVER TIME. .............................................................. 19

CHAPTER 2
FIGURE 2.1. SERIAL TRANSPLANTATION EXPERIMENTAL OVERVIEW.............................................. 29
FIGURE 2.2. CLONAL ABUNDANCE CHANGES BETWEEN PRIMARY AND SECONDARY
RECIPIENTS. ................................................................................................................ 30
FIGURE 2.3. CLONAL ABUNDANCE COMPARISON DURING SERIAL TRANSPLANTATION. .................. 31
FIGURE 2.4. LEUKEMIA CLONAL EXPANSION IN PERIPHERAL BLOOD OVERTIME. ............................ 33
FIGURE 2.5. SERIAL TRANSPLANTATION EXPERIMENTAL DESIGN. .................................................. 34
FIGURE 2.6. LEUKEMIA CLONE CLUSTERS BASED ON BEHAVIOR DURING SERIAL
TRANSPLANTATION. .................................................................................................... 35
FIGURE 2.7. ALL GENES TESTED IN COMPARING CLONES THAT EXPANDED AND DIMINISHED
DURING SERIAL TRANSPLANTATION. ........................................................................... 36
FIGURE 2.8. CO-TRANSPLANTATION EXPERIMENTAL DESIGN. ........................................................ 37
FIGURE 2.9. CLONAL ABUNDANCE IN CO-TRANSPLANTATION EXPERIMENTS. ................................ 38

CHAPTER 3
FIGURE 3.1. CLONAL ABUNDANCE ACROSS DIFFERENT TISSUES AND ORGANS. .............................. 48
FIGURE 3.2. CLONAL DISTRIBUTION ACROSS DIFFERENT TISSUES AND ORGANS. ............................ 49
ix
FIGURE 3.3. BRIDGING BARCODE CLONAL BEHAVIOR WITH SINGLE-CELL RNA SEQUENCING
DATA. ......................................................................................................................... 50
FIGURE 3.4. DIFFERENTIALLY EXPRESSED GENES BETWEEN BLOOD AND BONE MARROW. ............. 51
FIGURE 3.5. EXTRAMEDULLARY EXPANSION IN PATIENT SAMPLE JFK93 NAÏVE. ........................... 52
FIGURE 3.6. EXTRAMEDULLARY EXPANSION IN PATIENT SAMPLE JFK88 NAÏVE. ........................... 53
FIGURE 3.7. EXTRAMEDULLARY EXPANSION IN PATIENT SAMPLE ALL04. .................................... 54
FIGURE 3.8. DIFFERENTIALLY EXPRESSED GENES BETWEEN BLOOD AND OVARY. .......................... 55

CHAPTER 4
FIGURE 4.1. CHEMOTHERAPY TREATMENT GROUPS. ...................................................................... 75
FIGURE 4.2. EXPERIMENTAL MODEL RECAPITULATES CLINICAL ALL TREATMENT. ....................... 76
FIGURE 4.3. CLONAL ABUNDANCE BEFORE AND AT END POINT OF TREATMENT. ............................ 77
FIGURE 4.4. CLONAL DYNAMICS DURING INTENSIVE + MAINTENANCE CHEMOTHERAPY
TREATMENT. ............................................................................................................... 77
FIGURE 4.5. COMPARING CHEMOTHERAPY RESPONSE OF THE SAME LEUKEMIA CLONES IN
DIFFERENT TISSUES. .................................................................................................... 78
FIGURE 4.6. ALL CLONES DIFFERENTIALLY RESPOND TO CHEMOTHERAPY TREATMENTS. ............. 79
FIGURE 4.7. ALL04 CLONES THAT RESPONDED SIGNIFICANTLY BETTER TO INTENSIVE +
MAINTENANCE THERAPY THAN TO INTENSIVE THERAPY. ............................................ 80
FIGURE 4.8. ALL04 CLONES THAT RESPONDED SIGNIFICANTLY BETTER TO MAINTENANCE
THERAPY THAN TO INTENSIVE THERAPY. .................................................................... 81
FIGURE 4.9. ALL20 CLONES THAT RESPONDED SIGNIFICANTLY BETTER TO INTENSIVE
THERAPY THAN TO MAINTENANCE THERAPY. ............................................................. 82

x
4 LIST OF SUPPLEMENTARY FIGURES
CHAPTER 1
SUPPLEMENTAL FIGURE 1.1 FACS GATING FOR HUMAN B-ALL CELLS IN THE PERIPHERAL
BLOOD. ....................................................................................................................... 20
SUPPLEMENTAL FIGURE 1.2. CLONAL DIVERSITY CHANGE IN THE PERIPHERAL BLOOD OVER
TIME. .......................................................................................................................... 21
SUPPLEMENTAL FIGURE 1.3. PERCENT OF HUMAN CELLS THAT ARE GFP POSITIVE. ...................... 21
SUPPLEMENTAL FIGURE 1.4. CLONAL DYNAMICS OF HUMAN B-ALL IN THE MOUSE BLOOD. ........ 22

CHAPTER 3
SUPPLEMENTAL FIGURE 3.1. CLONAL DISTRIBUTION ACROSS DIFFERENT TISSUES AND
ORGANS FROM ALL PRIMARY MICE. ............................................................................ 59
SUPPLEMENTAL FIGURE 3.2. CLONAL DISTRIBUTION ACROSS DIFFERENT TISSUES AND
ORGANS FROM SECONDARY RECIPIENT MICE. ............................................................. 61
SUPPLEMENTAL FIGURE 3.3. IMAGES OF ALL MICE THAT WERE XENOGRAFTED WITH THE
HUMAN B-ALL SAMPLES. .......................................................................................... 62
SUPPLEMENTAL FIGURE 3.4. CLONAL ABUNDANCE ACROSS DIFFERENT TISSUES AND ORGANS
OF MICE THAT WERE XENOGRAFTED WITH PATIENT SAMPLE ALL04. ......................... 63

CHAPTER 4
SUPPLEMENTAL FIGURE 4.1. HUMAN CHIMERISM DURING CHEMOTHERAPY. ................................. 83
SUPPLEMENTAL FIGURE 4.2. CLONAL DYNAMICS IN ALL20 XENOGRAFTS DURING
CHEMOTHERAPY. ........................................................................................................ 85
SUPPLEMENTAL FIGURE 4.3. CLONAL DYNAMICS IN ALL06 XENOGRAFTS DURING
CHEMOTHERAPY. ........................................................................................................ 87
SUPPLEMENTAL FIGURE 4.4. CLONAL DYNAMICS IN ALL04 XENOGRAFTS DURING
CHEMOTHERAPY. ........................................................................................................ 89
SUPPLEMENTAL FIGURE 4.5. QPCR ANALYSIS OF GFP NEGATIVE CLONES. .................................. 90
xi
SUPPLEMENTAL FIGURE 4.6. ALL04 CLONES THAT RESPONDED DIFFERENTLY UNDER
DIFFERENT CHEMOTHERAPY TREATMENTS. ................................................................ 91
SUPPLEMENTAL FIGURE 4.7. ALL06 CLONES THAT RESPONDED DIFFERENTLY UNDER
DIFFERENT CHEMOTHERAPY TREATMENTS. ................................................................ 92
SUPPLEMENTAL FIGURE 4.8. ALL20 CLONES THAT RESPONDED DIFFERENTLY UNDER
DIFFERENT CHEMOTHERAPY TREATMENTS. ................................................................ 93
SUPPLEMENTAL FIGURE 4.9. ALL06 CLONES THAT RESPONDED SIGNIFICANTLY BETTER TO
INTENSIVE + MAINTENANCE TREATMENT THAN TO INTENSIVE TREATMENT................ 93
SUPPLEMENTAL FIGURE 4.10. ALL GENES TESTED IN ALL04 CLONES THAT RESPONDED
SIGNIFICANTLY BETTER TO THE INTENSIVE + MAINTENANCE THERAPY THAN
TO THE INTENSIVE THERAPY COMPARED WITH ALL OTHER ALL04 CLONES. ............... 94
SUPPLEMENTAL FIGURE 4.11. ALL GENES TESTED IN ALL04 CLONES THAT RESPONDED
BETTER TO MAINTENANCE THERAPY THAN TO INTENSIVE THERAPY COMPARED
WITH ALL OTHER ALL04 CLONES. .............................................................................. 94
SUPPLEMENTAL FIGURE 4.12. ALL GENES TESTED IN ALL20 CLONES THAT RESPONDED
BETTER TO INTENSIVE THERAPY THAN TO MAINTENANCE THERAPY COMPARED
WITH ALL OTHER ALL20 CLONES. .............................................................................. 95

xii
5 LIST OF TABLES
CHAPTER 1
TABLE 1.1. PATIENT SAMPLE INFORMATION .................................................................................. 19

CHAPTER 2
TABLE 2.1 PEARSON CORRELATION VALUES OF CLONAL ABUNDANCE BETWEEN DIFFERENT
MICE DURING SERIAL TRANSPLANTATION. ................................................................... 39

CHAPTER 3
TABLE 3.1. PEARSON CORRELATION VALUES FROM THE COMPARISON OF BLOOD, SPLEEN
AND BONE MARROW FROM SPINE.................................................................................. 56
TABLE 3.2. PEARSON CORRELATION VALUES FROM THE COMPARISON OF BONE MARROW
FROM DIFFERENT ANATOMICAL SITES. ......................................................................... 57

APPENDIX
TABLE A.1. MONOCLONAL ANTIBODIES USED IN THIS STUDY. ..................................................... 102

1
1 CHAPTER 1

CLONAL TRACKING OF HUMAN ACUTE LYMPHOBLASTIC
LEUKEMIA USING PATIENT DERIVED XENOGRAFTS

1.1 Abstract
The current understanding of tumor clonal dynamics is largely derived from next-
generation sequencing of bulk tumor genomes (1–3) However, many of these studies have relied
on naturally occurring genetic mutations (4, 5). These mutations are established at different times
therefore, the clones carrying them cannot be directly compared to assess cellular interactions. To
track the heterogeneous activities of individual cancer cells and study cellular interactions, we
applied a genetic barcoding technology that labels individual cells with a small segment of DNA
(6). The DNA barcodes are inserted into the cellular genomes and inherited by all descendants of
the barcoded cells. Activities of barcoded cells can be evaluated using patient-derived xenograft
(PDX) models. In this study, we label patient B-cell acute lymphoblastic leukemia (B-ALL) cells
prior to transplantation into immune deficient mice and establish a B-ALL PDX model. We studied
leukemia cells acquired from patients at different disease stage and demonstrate that clonal
dynamics differ in vivo. Finally, we show that clonal abundance distribution gradually spreads out
overtime with the most abundant clones becoming more abundant.

2
1.2 Introduction
Advances in cancer clonal tracking
Clonal evolution of cancer has been extensively studied using bulk tumor cell populations.
Next generation sequencing (NGS) techniques such as whole genome or whole exome sequencing
have been generally used to track variant allele frequencies within tumors (7–11). However, with
bulk samples, DNA from thousands or millions of cells is mixed prior to sequencing. Therefore,
the result is generally an estimation of the different variants within the sample, with no precise
information of the evolutionary history of the tumor. Sophisticated computational analysis such as
PyClone (12), SciClone (13) and PhyloWGS (14) have been created in order to infer clonal
frequencies from these bulk sequencing approaches, however they only provide a snapshot of the
mutational landscapes at certain time points (e.g. diagnosis and relapse). Nonetheless, these studies
have been instrumental in establishing the existence of clonal heterogeneity (15, 16), identifying
certain driver mutations (17, 18) and reconstructing clonal architecture (15, 19, 20).
Single-cell sequencing has offered an alternative approach as it involves single-cell
isolation, and then genomic amplification, followed by sequencing. With single-cell studies, new
details about clonal frequencies and their evolution during tumor progression has emerged. For
example, genetically distinct clones with unique biological and clinical properties have been
identified in acute lymphoblastic leukemia (ALL) and colon cancer (21, 22). Single-cell SNV
(single-nucleotide variants) analysis of patients with ALL revealed co-dominant clones and
showed that KRAS mutations occur late in disease development but are not sufficient for clonal
dominance (21). In colon cancer, a dominant cancer clone was found to possess APC and TP53
mutations typical of colon cancers, but a rare subclone with CDC27 and PABPC1 mutations was
also identified (22). Furthermore, single-cell sequencing has also provided insight into the debate
3
of whether drug resistance is caused by the selection of rare pre-existing clones or through acquired
resistance by induction of new mutations. Single-cell DNA sequencing of samples from twenty
patients with breast cancer before and after chemotherapy showed that resistant genotypes are pre-
existing and selected by chemotherapy (23).
Altogether, significant amount of work has been done to elucidate the clonal heterogeneity
and the evolution of cancer. However, much remains unknown and the answers are hindered by
current technique limitations such as the resolution limits of NGS.

Genetic barcodes in cancer clonal tracking
Regardless of recent improvements in sequencing techniques most of these studies rely on
naturally occurring genetic mutations (4, 5). Since mutations are established at different times, the
clones carrying them cannot be directly compared to assess cellular interactions. To overcome this
limitation in cancer clonal tracking, recent efforts have increasingly favored DNA barcoding. This
technique is a lineage tracing approach that introduces a unique heritable marker into the genome
of individual cells such that they and their descendants can be tagged and tracked over time (6, 24,
25). Using this approach, it was found that human breast tumor subclones exhibit diverse growth
kinetics in serially passaged xenografts (25). Barcoding revealed a shared and independent ability
of basal cells and luminal progenitors, isolated from normal human mammary tissue and
transduced with a single oncogene, to produce serially transplantable, polyclonal, invasive ductal
carcinomas (26). In addition, groups have identified the presence of pre-existing and de novo
mutations that can respectively drive resistance to therapy (27, 28). Chemotherapy treatment in
4
PDX models derived from patients with triple negative breast cancer, was found to have minor
impact on the barcode clonal diversity in relapsed tumor (29).

Genetic barcode experimental system
In this study, we employ an in vivo clonal tracking methodology developed in our lab. The
efficacy and the sensitivity of this tracking system have been demonstrated in several research
articles (6, 30–32). This system can quantify cellular proliferation over long periods of time while
preserving the integrity of the whole cell population and maintaining cellular interactions. These
unique features allow clonal dynamics of cancer cells to be measured at both the cellular and
molecular levels under physiological conditions. This experimental system utilizes synthesized
DNA segments to uniquely label individual cells. The DNA segments are incorporated into the
cellular genome and serve as genetic ‘barcodes.’ They are not expressed but are inherited by
progenies along with genomic DNA. Therefore, the abundance of a genetic barcode in a cell
population is proportional to the number of progeny that the original barcoded cell produces. Using
this methodology, it was demonstrated that cellular interactions between hematopoietic stem cells
significantly alter their proliferation and differentiation (31). The system allows for the
quantification of cellular proliferation and competition with single-cell resolution over long
periods of time.

5
1.3 Results
Engraftment of patient derived leukemia
In order to establish a reliable and robust patient derived xenograft (PDX) from primary
cancer samples. A characterization of how the cancer cells engraft and the timing of the disease
progression is essential. In addition, for genetic barcode delivery it was imperative to determine
early whether the primary patient cells would be susceptible to lentiviral transduction. Several
groups have previously reported the feasibility of transducing primary leukemia cells using HIV-
1 derived lentiviral vectors with good efficiency and to a variable degree for different patient
samples (33–35). In addition, I have previously validated the use of luciferase bioluminescence
technology to confirm engraftment and evaluate treatment efficacy of primary patient cancer cells
(36) and cancer cell lines (37). Therefore, I generated a luciferase-eGFP encoding lentivirus to
address our primary concerns. Using this experimental system, I successfully demonstrated that
our primary human B-ALL cells are susceptible to lentiviral transduction and engraft in immune
deficient mice (Fig. 1.1). The data also illustrates that leukemia clones engraft and expand into
different locations in mice.

Cell-dose optimization
To examine the timing of leukemia engraftment and disease progression we transplanted
primary B-ALL cells from a human patient at four different doses: (i) 25,000, (ii) 50,000, (iii)
100,000 and (iv) 250,000 cells into irradiated NOD SCID gamma (NSG) mice via tail vein
injection. Five mice were used for each cell-dose transplantation. Leukemia cells were detected in
mouse peripheral blood as early as 4 weeks post transplantation in mice that received 250,000
cells. Leukemia cells were detected at week 6 in mice that received both 50,000 and 100,000 cells.
6
It took 8 weeks post transplantation for leukemia cells to be detected in the peripheral blood of
mice that were transplanted with 25,000 (Fig. 1.2). At week 12 post transplantation the 25,000 and
50,000 cell-dose remained closely similar in percent engraftment at 84-86%. The mice that were
transplanted with 250,000 cell-dose exhibited high engraftment (93%) and had to be euthanized
shortly after due to the disease burden. It was concluded that 100,000 cell-dose would be ideal for
future experiments. This cell dose would be feasible for lentiviral transduction and provide enough
time during disease progression to monitor and recover barcoded leukemia cells.

Establishing a barcoded patient derived leukemia xenograft
We genetically barcoded individual primary human B-ALL cells derived from five patients
at different disease stages using a GFP-encoding lentiviral vector (Fig. 1.3A and Table 1.1). After
the barcode labeling, we transplanted the cells into sub-lethally irradiated NSG or NSG-SGM3
mice (Fig. 1.3A). In this study, we use the term ‘clone’ to specifically refer to the cells carrying
identical barcodes. After transplantation, we analyzed the peripheral blood of the recipient mice at
multiple time points. At the end time point, we analyzed cells from the perfused peripheral blood,
the spleen, the bone marrow, and any organs with noticeable metastasis. In serial transplantation
experiments, we sorted barcoded cells from the spleen (hCD45+/hCD19+/GFP+, fig. S1.1) and
transplanted them to the secondary recipients and subsequently to the tertiary recipients. Our data
show that barcoded leukemia cell population expanded proportionally to non-barcoded leukemia
cell population (Fig. 1.3B), suggesting that data from barcoded cells can represent the overall
leukemic cell population.
7
In vitro growth phenotypes of B-ALL cells derived from different disease stages
To compare clonal dynamics for samples derived from different disease stages, we
obtained matching human B-ALL samples from the same patients, a diagnosis (naïve) and a
treatment relapsed sample from three patients (JFK86, JFK88 and JFK93) (Fig. 1.3A). Equal cell
numbers were plated onto a 96-well plate for lentiviral transduction. (Fig. 1.4) Cells underwent a
24-hour stimulation period prior to adding the barcode encoding lentiviral vectors. Cells were
imaged in vitro at the 24-hour (Fig. 1.4A) and 48-hour (Fig. 1.4B) time-period coinciding with
before and after the addition of the lentivirus. Noticeable in vitro growth phenotype differences
were observed. First, patient JFK86 and JFK88 patient cells appeared to grow in clusters whereas
patient JFK93 did not. Second, patient JFK93 appeared to be fast-growing whereas JFK88 and
JFK86 demonstrated a slower-growing phenotype. Together, these data elucidate the differences
found across patients and between different disease stages.

In vivo clonal dynamics of B-ALL cells derived from different disease stages
To compare in vivo clonal dynamics for samples derived from different disease stages, we
employed our genetic barcode tracking technology to the B-ALL cells. We attempted to barcode
label the samples from the three patients, however patient JFK86 was not susceptible to lentiviral
transduction (data not shown). After barcode labeling and xenotransplantation, human leukemia
cells grew quickly and out-competed the endogenous mouse cells (Fig. 1.5A). At the end time
point, almost all mononucleated cells (MNC) in the peripheral blood were descendants of the
human leukemia cells. Noticeably, the samples from patient JFK93 grew much faster than those
from patient JFK88, this is consistent with our in vitro observations. Comparing the fast-growing
samples from patient JFK93, naïve cells grew much faster than relapsed cells in all experimental
8
mice (Fig. 1.5A). But this difference was not seen in the slow growing samples from patient JFK88.
These data demonstrated variations between disease stages and between patients.
To determine how many donor cells engrafted in the PDX model, we counted the barcode
numbers in all examined tissues and organs and analyzed the fractions of barcoded cells among all
donor cells. We found that less than 1% donor cells engrafted (Fig. 1.5B). This suggests that the
PDX model assesses a very small fraction of the cells. Similar engraftment barrier is also
commonly observed in mouse-to-mouse transplantation of our other studies (30). Comparing
samples from the same patient at different disease stages, on average more barcodes were detected
in the peripheral blood from the naïve samples than from the relapsed samples (Fig. 1.5C, for both
JFK88 and JFK93: P value < 0.01 at the initial time points; P value < 0.001 across all time points).
The clonal diversity was also much lower for the relapsed samples compared with the naïve
samples (fig. S1.2). Naïve B-ALL cells from patient JFK88 were significantly higher transduced
than the relapsed cells from the same patient under the same transduction condition (fig. S1.3),
which is partially responsible for the barcode number differences.

Low abundance clones are eliminated over time
Over time, the number of leukemia clones generally decreased for the slow-growing
samples derived from patient JFK88 (Fig. 1.5C) This reduction was caused by the elimination of
low abundance clones (Fig. 1.5D, Fig. 1.6, and fig. S1.4). While the growth of high abundance
clones was generally stable (Fig. 1.6 and fig. S1.4), clonal competition was evident during the later
disease stages when the overall growth slowed down (Fig. 1.5A, Fig. 1.7, and fig. S1.4). In most
mice that received the JFK93 relapsed sample, there was exactly one clone that outcompeted other
9
clones (Fig. 1.7). No significant differences were observed between the naïve and relapsed samples
during the dynamic clonal expansion in PDX mice, although clonal dominance was greater in mice
receiving relapsed samples than those receiving naïve samples.

1.4 Discussion
Understanding the clonal dynamics of cancer cells is essential for our understanding of
leukemia pathogenesis and progression. Here, we describe the tracking of leukemia clonal
dynamics of patient-derived B-ALL xenografts using genetic cellular barcodes. This genetic
barcoding system offers several advantages over previously used methods of clonal tracking
analysis: it is relatively simple, quantitative, and it is high throughput. In addition, as this system
does not rely on naturally occurring mutations, we can address questions of cancer clonal evolution
that cannot be addressed using bulk genome sequencing techniques.
In this chapter we established a PDX model using leukemia cells derived from different
disease stages (naïve and relapsed). We optimize the cell dose for leukemia transplantation to allow
enough time to monitor leukemia progression in mice. We incorporate a genetic barcoding
technique to track patient derived leukemia clones in vivo and we demonstrate that patient-derived
leukemia xenografts differ in clonality based on the disease stage of the samples.
While an advantage of the PDX model is that it can be generated with a limited amount of
material. This model becomes questionable when the studied tumor type is particularly
heterogeneous. For example, within one tumor or metastasis, multiple subclones can exist, each
harboring different genetic and/or epigenetic alterations. Our study demonstrates that genetic
barcoding and xenotransplantation is a simple and reliable technique to address these concerns.
10
We can model the clonal evolution and to study the clonal landscape and dynamics of human
leukemia cells.
In future studies, we can use this experimental system to track the cancer clonal diversity
in serial transplantations. This would provide insights into the cancer clonal selection that occurs
during serial transplantations in PDX models. This is an important consideration that needs to be
addressed when testing cancer therapies using late passaged PDX. Furthermore, combining genetic
barcoding with in vivo chemotherapy would provide a powerful tool to study the clonal evolution
of chemotherapeutic resistance.

11
1.5 Figures

Figure 1.1. Xenograft model for studying human leukemia in mice.
In vivo luciferase images at various time points illustrate the location of human leukemia cells after
transplantation into irradiated immunocompromised mice. Color scale measures radiance in
(p/sec/cm2) which refers to the number of photons per second that are leaving a square centimeter
of tissue.

12

Figure 1.2
x 5 x 5 x 5 x 5
13
Figure 1.2. Circulating leukemia cells are detected earlier with higher cell dose
transplantation.
Leukemia cells from one patient were transplanted at different cell doses into irradiated
immunocompromised mice via tail vein injection. Five mice were injected with the same cell dose.
Included, are representative FACS plot for each cell dose transplantation.

14

Figure 1.3. Clonal tracking of human B-ALL cells using genetic barcodes.
(A) Human B-ALL cells acquired from three patients at distinct disease stages (chemotherapy
naïve and chemotherapy relapsed) are genetically labeled with DNA barcodes, using a GFP
encoding lentiviral vector, and transplanted into irradiated immunocompromised mice. Leukemia
engraftment is monitored through peripheral blood analysis. Cells are sorted during each blood
collection for barcode analysis. (B) The fractions of mouse cells, human B-ALL non-barcoded
cells (GFP-) and human B-ALL barcoded cells (GFP+) in the mononucleated cells (MNC) of the
peripheral blood over time.

15

Figure 1.4
16
Figure 1.4. Leukemia in vitro cultures.
Equal number of leukemia cells acquired at different disease stages (naïve (N) and relapsed(R))
from three patients (JFK86, JFK88, and JFK93) were cultured in vitro during the lentiviral
transduction and barcode delivery. (A) 24-hours after plating. (B) 48-hours after plating. White
bar represents a 10 μm scalebar.

17

Figure 1.5. Leukemia cells acquired at different disease stages.
(A) Human chimerism of the peripheral blood following transplantation of diagnostic (naïve) and
relapsed leukemia cells from the same patients. Each line depicts data from a single mouse
recipient. (B) Engraftment of human B-ALL cells. The engraftment rate was calculated based on
the number of donor cells and the number of clones detected from the blood, bone marrow, spleen
and enlarged extramedullary organs, adjusted for GFP%. Each dot represents data from one mouse
and the horizontal line depicts the mean. (C) Number of unique DNA barcodes detected in the
peripheral blood. Each line represents data from one mouse. (D) Clonal abundance distribution
over time. Histogram shows data from all mice.

18

Figure 1.6. Leukemia clones expand proportionally in PDX model.
Clonal composition in the peripheral blood from naïve and relapsed samples of the same patients
over time. Each color represents one distinct genetic barcode corresponding to a leukemia clone.
The sizes of each colored columns indicate their relative abundance. Additional mice are shown
in fig. S1.4.

19

Figure 1.7. Clonal abundance changes over time.
Each line represents one clone. Each color depicts one mouse. Shown are all clones exceeding 1%
MNC abundance at any time point.

Table 1.1. Patient sample information

20
1.6 Supplementary figures

Supplemental Figure 1.1 FACS gating for human B-ALL cells in the peripheral blood.
Shown is a representative peripheral blood analysis of a mouse transplanted with barcoded (GFP+)
and unbarcoded (GFP-) B-ALL cells six weeks post transplantation.

21

Supplemental Figure 1.2. Clonal diversity change in the peripheral blood over time.
Each line represents data from one recipient mouse. * P<0.05.

Supplemental Figure 1.3. Percent of human cells that are GFP positive.
Each line represents data from one recipient mouse. ***P<0.001.

22

Supplemental Figure 1.4. Clonal dynamics of human B-ALL in the mouse blood.
Shown are all experimental mice in addition to those in Fig. 1.6 that received samples from naïve
(A, C) and matched relapsed JFK88 and JFK93 (B, D) patient samples respectively. Each color
represents one distinct genetic barcode corresponding to a leukemia clone. The sizes of each
colored columns indicate their relative abundance.

23
2 CHAPTER 2

ACUTE LYMPHOBLASTIC LEUKEMIA CLONAL SELECTION DURING
SERIAL TRANSPLANTATION

2.1 Abstract
The patient derived xenograft (PDX) has become an indispensable model for studying
human cancer biology. It has evolved as a particularly robust system as it can be generated with
limited amount of material and generally early passage PDXs maintain high degree of genetic
fidelity compared to their original tumor (38, 39). In this chapter, we study the clonal diversity in
PDXs established from B-cell acute lymphoblastic leukemia patient cells. We utilize a genetic
barcode tracking system to follow leukemia clones in vivo and compare PDX mice from serial
transplantations. We demonstrate that leukemia clonal heterogeneity is exacerbated by PDX serial
transplantation. This is particularly troubling when late passage PDX are used to test therapeutic
response of potentially new cancer drugs. In addition, barcode clonal abundance data demonstrate
that leukemia clones either expand or diminish during serial transplantation. Lastly, we find that
variability in clonal expansion is largely cell autonomous and associated with differential gene
expression.

24
2.2 Introduction
Cancer clonal evolution
Leukemia is thought to arise and evolve in a Darwinian type of clonal evolution. In which,
cancer results from the accumulation of multiple mutations in a single-cell lineage. They are
sequentially acquired and then subject to an evolutionary process where selection drives the
expansion of more ‘fit’ subclones. In acute lymphoblastic leukemia (ALL), many aspects of clonal
evolution are still subject to ongoing debate. For instance, the existence of leukemia stem cells,
their frequency within the total leukemic population, and the dynamics of their clonal offspring
remain unknown. Many aspects of clonal evolution are poorly understood, and answers are
currently hindered by the technical challenge of distinguishing and isolating distinct cancer
subclones at the single-cell level.

Patient derived xenograft model
Patient derived xenografts (PDXs) established directly from primary tumor tissue is a
superior model for studying cancer biology as these models more closely resemble the human
microenvironment and are not compromised by in vitro adaption of cancer cells. PDXs harbor
bona fide tumor targets directly from patients, and hence their use in drug discovery has exploded.
For example, a study in sarcoma tumors showed that many copy number alteration changes found
in PDX are frequently observed in patients (40). In breast cancer, many of the mutations detected
in the PDX were also observed in brain metastases derived from the same patient (41). These
studies are just a few of many that has made the PDX experimental model indispensable in
translational cancer research. Nonetheless, these studies have been done using bulk samples. A
25
single-cell analysis is necessary to fully characterize cancer clonal dynamic during serial
passaging.

Cancer clonality through serial transplantation
Cancer clonal dynamics through serial transplantations has been increasingly studied. In
colon cancer, by combining DNA copy number alteration (CNA) profiling, sequencing, and
lentiviral lineage tracking, the repopulation dynamics of 150 single lineages from 10 human
colorectal cancers were followed and analyzed through serial passages in mice (42). This analysis
distinguished individual clones and showed that clones remained stable upon serial transplantation.
In breast cancer, DNA barcodes were used to lineage trace cell line derived clones. It was
determined that clone number decreased during passage (25). An extremely comprehensive
analysis was done monitoring CNAs in 1,110 PDX samples across 24 cancer types (39). This study
concluded that that despite an overall similarity, the CNA landscapes of PDXs diverge
substantially from their parental tumors during passaging (39). While these studies have been
important to elucidate the limitations in PDX when utilizing late passaged tumor samples, an
understanding of underlying molecular mechanisms remain unknown.

26
2.3 Results
Clonal selection occurs during serial transplantation
Most PDX studies use mice after multiple passages in order to obtain enough biological
replicates. To determine how clonal diversity is influenced by transplantation, we performed serial
transplantations using barcoded leukemia clones derived from two different patients, ALL04 and
JFK93 (Fig. 2.1). We found that some clones abundant in primary recipient mice were not detected
in the secondary recipients (Fig. 2.2 and Fig. 2.3A). Clonal correlations between secondary
recipients were much greater than the correlations between the primary and secondary recipients
(Fig. 2.3B and table S2.1). Leukemia clones that engrafted in secondary recipients expanded
proportionally in the mouse peripheral blood in secondary recipients’ overtime (Fig. 2.4).

Leukemia clonal behaviors during serial transplantation is associated with gene expression
differences
We clustered the leukemia clones based on their abundance in the primary, secondary and
tertiary recipients and identified clones that expand or diminish during the serial transplantation
(Fig. 2.5 and Fig. 2.6). We also performed single-cell RNA sequencing using the donor cells of
the tertiary recipients. Because our tracking barcodes are transcribed, we were able to map the
single-cell RNA sequencing data with the barcode clonal tracking data. We found five genes
including HSPH1, IKZF2, MT-ND3, RBM3, and UGP2, that were significantly differentially
expressed between the expanding clones and the diminishing clones derived from two patients
(Fig. 2.6 and Fig. 2.7). UGP2 and RBM3 are downregulated in expanding clones. IKZF2, HSPH1
and MT-ND3 are upregulated in expanding clones. Some of these genes have been previously
shown to regulate cancer expansion. Downregulation of RBM3 (RNA-binding motif protein 3) is
27
associated with melanoma progression (43). IKZF2 (IKAROS Family Zinc Finger 2) supports
leukemogenesis of leukemia stem cells in acute myeloid leukemia (AML) (44). HSPH1 (Heat
Shock Protein Family H (Hsp110) Member 1) regulates the growth of human B-cell non-Hodgkin
lymphoma and has been proposed as a therapeutic target (45). These data demonstrated how clonal
diversity changes with transplantation and suggested that the changes were driven by cell intrinsic
factors, which may reflect the adaptability of human leukemia cells to the mouse environment.

2.4 Discussion
The ability to directly transfer human tumor cells into mice, and propagate them for
multiple passages in vivo, offers unique opportunities for cancer research and drug discovery. This
has made PDXs an indispensable cancer model. Like any other model system, however,
understanding its limitations and the ways in which it differs from primary human tumors in their
natural environment is required for optimal application. The observation that leukemia clonal
diversity is reduced during serial transplantation (25, 39, 42) has raised the question as to which
extent functional differences between individual leukemia clones dictate their survival and/or drive
their selection. Here, we demonstrate that variability in clonal expansion is largely cell autonomous
and is associated with differential gene expression.
We used a k-means clustering approach and observed two main types of clonal behavior
during serial transplantation, expanding and diminishing. Similarly, in colon cancer three distinct
behaviors observed in cancer clones were identified. They were; (i) type I or persistent clones, (ii)
type II or short-term clones and (iii) type III or transient clones (42). Together our findings support
the idea that substantial functional diversity with respect to clonal longevity during serial
propagation exists.
28
We identified UGP2 and RBM3 to be significantly downregulated in expanding clones.
RBM3 (RNA-binding motif protein 3), is a glycine rich protein containing a RNA-recognition
motif (RRM) through which it binds to both to DNA and RNA (46). In contrast to our observations,
RBM3 downregulation in colon cancer cell lines decreases cell growth in culture (47). Differences
can be attributed to the type of cancer and/or to the in vitro adaption of cancer cells. Nonetheless,
it plays a role in cancer progression.
IKZF2, HSPH1 and MT-ND3 are upregulated genes in expanding clones. Some of these
genes have also previously been found to regulate cancer expansion. Consistent with our results,
that IKZF2 AND HSPH1 may support growth in ALL expanding clones is in accordance with
recent published data. IKZF2 depletion in AML reduced colony formation, increased
differentiation and apoptosis, and delayed leukemogenesis (44). HSPH1 was found to promote
stabilization of key lymphoma oncoproteins and silencing of HSPH1 was associated with a
significant growth delay and tumorgenicity in B-cell non-Hodgkin lymphoma (45).
In conclusion, our data reveals novel genes that may be involved in regulating cancer
progression (MT-ND3 and UGP2). UGP2 (UDP-glucose pyrophosphorylase 2) plays a central role
as a glucosyl donor in cellular metabolic pathways and MT-ND3 (Mitochondrially Encoded
NADH:Ubiquinone Oxidoreductase Core Subunit 3) related pathways are respiratory electron
transport. These genes point to a role of cellular metabolism.

29
2.5 Figures

Figure 2.1. Serial transplantation experimental overview.
Human leukemia cells were barcoded before primary transplantation and transplanted into multiple
secondary recipients.

30

Figure 2.2. Clonal abundance changes between primary and secondary recipients.
Clonal abundance changes during serial transplantation. Data were collected from the spleen and
normalized among barcoded cells. Five independent experiments were performed using samples
from two patients, (A) JFK93 and (B) ALL04. Each color within a column represents one distinct
genetic barcode corresponding to a leukemia clone. Each column shows data from one mouse.

31

Figure 2.3. Clonal abundance comparison during serial transplantation.
(A) Clonal abundance comparison between the primary and secondary recipients during serial
transplantation. Data were collected from the spleen and normalized among barcoded cells. Left
graph includes data from two independent experiments using patient ALL04. Right graph includes
data from three independent experiments using patient JFK93 naïve. Each dot represents a
barcoded clone, and each color depicts one primary mouse. (B) Pearson correlation of clonal
abundance between different mice during serial transplantation. Pearson correlation values are
provided in table 2.1. ***P<0.001.

32

Figure 2.4.
33
Figure 2.4. Leukemia clonal expansion in peripheral blood overtime.
Clonal composition in the peripheral blood overtime of secondary recipient mice transplanted with
(A) JFK93 and (B) ALL04 leukemia cells. Each color represents one distinct genetic barcode
corresponding to a leukemia clone.

34

Figure 2.5. Serial transplantation experimental design.
Barcoded leukemia cells were transplanted through three generations of mice. Clonal abundance
was assessed in each recipient. Barcoded leukemia cells from the spleen of secondary recipients
were transplanted to the next recipients and were also analyzed using single-cell RNA sequencing.
Clonal abundance was assessed in each recipient (primary, secondary and tertiary) and mapped to
the single-cell RNA sequencing data as the genetic tracking barcodes were transcribed.

35

Figure 2.6. Leukemia clone clusters based on behavior during serial transplantation.
(A) K-means clustering separated leukemia clones into diminishing and expanding groups, after
low abundant clones were excluded, in two patient samples. (B) Five genes were identified as
significantly differentially expressed between diminishing and expanding clones using spleen cells
from the secondary recipients. UGP2 and RBM3 are downregulated in expanding clones. IKZF2,
HSPH1 and MT-ND3 are upregulated in expanding clones. The black bar indicates the mean, and
the white dot represents the median.

36

Figure 2.7. All genes tested in comparing clones that expanded and diminished during serial
transplantation.
Shown are False Positive score (FPS) and P values of differential expression. Each dot represents
a gene. Dashed lines are at 0.05 and 0.05 for FPS and P value. FPS calculation is described in the
method section.

37

Figure 2.8. Co-transplantation experimental design.
B-ALL cells from one patient were divided into three equal portions, labeled with different barcode
libraries, and transplanted into separate primary mice. Barcoded leukemia cells were recovered
from the primary mice and combined equally in the secondary transplantation

38

Figure 2.9. Clonal abundance in co-transplantation experiments.
Clonal abundance in secondary recipients of co-transplantation. Data from four independent
experiments using patient sample ALL04, ALL20, ALL06 and JFK93 is included. Each column
corresponds to an individual mouse. Each section in the column represents a clone. The color
illustrates which primary mouse the clones are derived from.

39
Table 2.1 Pearson correlation values of clonal abundance between different mice during
serial transplantation.

40
3 CHAPTER 3

ACUTE LYMPHOBLASTIC LEUKEMIA CLONAL DISTRIBUTION
ACROSS DIFFERENT TISSUES AND ORGANS

3.1 Abstract
Cancer is an evolving disease driven by genetic and epigenetic changes (42). These
molecular alterations generate tremendous cellular heterogeneity (42). Consequently, individual
cancer cells differentially proliferate and selectively metastasize. To address this, many studies
have investigated molecular differences between individual cancer cells using single-cell RNA
sequencing (42, 48–50). However, not much has been done to investigate the heterogeneous
activities of individual cancer cells and relate their activity differences to differences in their gene
expression. Comparing cancer cells that exhibit different levels of proliferation, would provide
unique opportunities to uncover the cellular and molecular mechanisms underlying these
processes. In this study, we utilize an experimental system that combines genetic barcoding with
high throughput sequencing to track leukemia clones in vivo using patient derived xenografts. The
leukemia clonal comparison between tissues and organs reveals marked anatomical biases, which
is surprising for a “liquid” tumor. Our findings provide important implications for clinical
diagnostic sampling.

41
3.2 Introduction
Acute lymphoblastic leukemia
Acute lymphoblastic leukemia (ALL) is a hematological malignancy defined by the clonal
expansion and accumulation of abnormal immature lymphoid precursors cells in the bone marrow
(BM) compartment. The abnormal cell population replaces normal hematopoietic cells and can
spread from the BM to the entire body through systematic circulation. Often this results in an
extramedullary expansion of leukemia cells. Interestingly, the site of relapse in ALL patients is
one of the most important prognostic factors (51). While ALL relapse occurs most often in the
BM, extramedullary relapses are also possible albeit less commonly reported. Unfortunately, the
molecular mechanism of how leukemia cells migrate and invade distant favorable micro-
environments from their native marrow niche is yet to be elucidated. A better understanding of
how leukemia cells circulate and how the microenvironment plays a role in their preferential
expansion will provide insights for treating patients who relapse following chemotherapy.

Acute lymphoblastic leukemia extramedullary expansion
Given that leukemia cells spread throughout the body via systemic circulation, ALL
involvement in extramedullary sites is common and can cause lymphadenopathy, splenomegaly or
hepatomegaly (52). Although less common, ovaries are also involved in ALL (53–56) as well the
kidney (56, 57). An autopsy study provided conclusive evidence that ALL can infiltrate and
expand in several tissues and organs including the major blood vessels, large intestines, prostate,
central nervous system (CNS), thymus, and ovaries, among others (53).
42
Cancer metastasis
Metastasis is the process of cancer cells spreading from the primary tumor to surrounding
tissue and distant organs and it is the primary cause of morbidity and mortality (58, 59). Several
studies have presented evidence indicating that tumors exhibit genetic heterogeneity both across
different anatomic regions (60–63) and within single cancer-tissue samples (41, 64–66). For
example, in pancreatic cancer, evidence presented indicated that there is genetic heterogeneity
among metastasis-initiating cells, that seeding metastasis may require driver mutations beyond
those required for primary tumors and that phylogenetic trees across metastases show organ-
specific branches (60). In addition, metastatic prostate cancers were found to have monoclonal
origins and to maintain a unique signature copy number pattern of the parent cancer cell (62).
These data attest to the abundance of genetic variation in cancer, brought about by genomic
instability and evolutionary selection.

CXCL12/CXCR4
CXCL12, also known as stromal derived factor 1 alpha (SDF1α), is one of the most
important chemokines in the BM niche (67). Through binding to its receptor, CXCR4, CXCL12
favors hematopoietic stem cell homing to the BM and participates in the maintenance of these cells
in the niche. CXCL12 has been shown to be involved in metastasis in many solid tumor cancer
models, such as breast, prostate, lung cancer, pancreatic adenocarcinoma, and others (68). In an
ovarian cancer model, the most advanced stages of the disease and most invasive phenotypes are
associated with increased CXCR4 expression at the cancer cell surface. More importantly, ascites
samples from patients with advanced ovarian cancer showed significantly higher levels of
CXCL12 (69). The CXCL12-CXCR4 axis is also involved in the pathogenesis of leukemia, as it
43
favors not only leukemia cell homing in the BM, but also their survival and proliferation (70–73).
The role of this chemokine axis has been implicated in ALL infiltration into the testis (74).
Together, these data suggest that ‘liquid’ and ‘solid’ tumor cells communicate with their
microenvironment using similar methods such as the CXCL12/CXCR4 chemokine axis.
Considering extramedullary infiltration of leukemias as metastases could be a useful approach to
get a better understanding of the molecular mechanisms driving leukemia circulation, infiltration
and homing.

3.3 Results
High leukemia clonal correlation between blood, spleen and bone marrow
To determine whether the clonal composition in the peripheral blood is representative to
that in the other organs in a patient derived xenograft (PDX) model, we calculated the Pearson
correlation of the clonal composition between different tissues and organs in primary recipient
mice of seven samples from five patients (Fig. 3.1A, 3.1B and Tables 3.1, 3.2). We found that
ALL clones in the blood and spleen were highly correlated in all mice except for two (ALL20-M1
and ALL20-M2) where one clone dominated in the blood but was not detected in other organs
(Fig. 3.1A and fig. S3.1A). In a few mice, clones in the bone marrow did not correlate well with
those in the blood and the spleen (Fig. 3.1A, Table 3.1). When we compared the bone marrow
from different anatomical sites, some mice exhibited high clonal correlation, while others showed
significant differences (Fig. 3.1B, Table 3.2). Mice that received ALL20 and JFK93N samples
mostly exhibited considerable differences (Fig. 3.1B, Table 3.2), suggesting that a cell intrinsic
mechanism is responsible for tissue homing patterns.

44
Spatially confined expansion of leukemia clones in the bone marrow
The disparate clonality in the bone marrow contradicts the prevalent dogma that leukemia,
a “liquid” cancer, uniformly spreads throughout the body (Fig. 3.1). Our data clearly demonstrate
that some clones were substantially expanded in the bone marrow from one anatomical site and
were not circulating (Fig. 3.2). This spatially confined clonal expansion was consistently detected
in 15 primary recipient mice that received B-ALL cells from 5 different patients (fig. S3.1). It was
not detected in any mice that received one particular patient sample (JFK88N) (fig. S3.1F),
suggesting that it is a cell autonomous characteristic. Spatially confined clonal expansion was less
detectable in secondary recipient mice (fig. S3.2) and completely missing in tertiary mice (data
not shown). This can be attributed to that fact that bone marrow expanded clones were under-
represented in the spleen where donor cells for serial transplantation were collected as by common
convention.

Leukemia clonal expansion in the bone marrow is associated with differentially expressed genes
Since our genetic tracking barcodes are transcribed, we are able to bridge clonal abundance
behavior to single-cell RNA sequencing data (Fig. 3.3) We compared gene expression of the clones
that were over and underrepresented in the bone marrow compared to the blood (Fig. 3.2B-3.2C),
and found three genes, LRIF1, BTK, and DNAJC1, that were significantly differentially expressed
(Fig. 3.4). BTK (Bruton’s tyrosine kinase) is critical for signal transduction downstream of pre-B
cell receptor (pre-BCR) and functions as a tumor suppressor in B-ALL (75, 76). A BTK-binding
molecule, Ibrutinib, is currently under clinical trial for treating B-ALL (ClinicalTrials.gov
Identifier: NCT02997761).
45
Extramedullary expansion of leukemia clones associated with differentially expressed genes
In addition to the clonal expansion in the bone marrow, we also found substantial
extramedullary clonal expansion in the stomach (Fig. 3.5), the kidneys (Fig. 3.6) and the ovaries
(Fig. 3.7) associated with three particular patient samples. All mice that received these patient
samples consistently exhibited enlargement of the respective organs (Fig. 3.5-3.7 and fig. S3.3).
We harvested human cells from these enlarged organs and compared their clones to the leukemia
clones found in the blood, spleen and bone marrow. We found that leukemia clones at the
extramedullary sites are often different from those in the hematopoietic tissues, including the
peripheral blood (Fig. 3.5-3.7 and fig. S3.1-S3.4). These data demonstrate clonal selection during
the extramedullary clonal expansion. We analyzed the donor cells and compared the gene
expression of the clones that expanded in the ovary and those that did not. We identified a largely
unknown gene, CMC2 (COX assembly mitochondrial protein 2 homolog), that was expressed
significantly higher in the clones that are overrepresented in the ovary (Fig. 3.8).

46
3.4 Discussion
Comparing extramedullary infiltration of leukemias as metastases of a liquid tumor can be
a useful approach for proposing more effective cancer therapies. Despite the clinical importance
of metastasis, fundamental questions about the clonal structures of metastatic tumors remain
unanswered. Here we harness the strengths of genetic barcoding and high throughput sequencing
to examine leukemia clonal behavior in mice xenografted with seven samples from five patients
with B-ALL. We combine this knowledge with single-cell RNA sequencing data and identify
genes that may play a role in leukemia circulation and/or infiltration.
The observation that leukemia clones in the peripheral blood and spleen are highly
correlated in xenografts, is not surprising given that the spleen is the largest lymphoid organ in the
body and plays an important role in immunological defenses. The primary function of the spleen
is to filter blood; therefore, technically any leukemia cell in circulation would essentially enter the
spleen.
We find that some leukemia clones were substantially expanded in the bone marrow from
one anatomical site and were not circulating and our single-cell RNA sequencing data identified
three genes (LRIF1, BTK, and DNAJC1) that were significantly differentially expressed. BTK has
been extensively studied in leukemia and found to be a tumor suppressor in B-ALL (75, 76). More
importantly, BTK is essential to chemokine-mediated homing and adhesion of B cells (77, 78).
Inhibition of BTK impairs BCR-controlled adhesion and CXCL12-controlled adhesion and
migration of chronic lymphocytic leukemia cells (79). Here, our in vivo data supports this report,
by demonstrating that leukemia cells with higher BTK expression are homing to the BM more
efficiently. Less in known about the other two identified genes, DNAJ1(DNAJ heat shock protein
47
family member C1) and LRIF1 (Ligand Dependent Nuclear Receptor Interacting Factor 1). Further
studies on these genes may provide insight into any therapeutic potential.
The asymmetry in spatial distribution within the bone marrow of leukemia clones has
important experimental and translational implications. This suggests that diagnostic patient
sampling of a single anatomic site may underestimate the clonal repertoire and may fail to obtain
certain and potentially relevant clones located elsewhere.
We observed disparate leukemia clonality between the hematopoietic tissues, including the
peripheral blood and the extramedullary sites. These data demonstrate clonal selection during the
extramedullary clonal expansion. Leukemia clones that are overrepresented in the ovary were
found to significantly express higher levels of CMC2 (COX assembly mitochondrial protein 2
homolog). This could be an interesting gene for studying the infiltration behavior of leukemia cells
into the ovary. It could prove to be a promising therapeutic target for the treatment of ovary
involvement in ALL relapsed patients.
It has been suggested that caution is required when inferring the genetic composition of
metastatic disease from a primary tumor biopsy, and vice versa. Indeed, our DNA barcode tracking
system reveals clonal differences in extramedullary infiltration of leukemia cells and single-cell
RNA sequencing data suggest that these clonal behaviors may be associated with gene expression
differences. In conclusion, our data presents relevant information that should be further
investigated as it may be crucial for patient diagnosis, treatment, and follow-up.

48
3.5 Figures

Figure 3.1. Clonal abundance across different tissues and organs.
Heatmaps showing Pearson correlations of clonal abundance across different tissues and organs.
Each column corresponds to one primary recipient mouse. (A) Comparison between the peripheral
blood (PB), spleen and bone marrow (BM) from spine. The clonal diversity in the blood of each
recipient is shown below where each color represents a unique clone. Pearson correlation values
are provided in Table 3.1. (B) Comparison of the BM harvested from three distinct anatomical
regions: left leg, right leg and spine. The clonal diversity in the spine of each recipient is shown
below. Pearson correlation values are provided in Table 3.2.
49

Figure 3.2. Clonal distribution across different tissues and organs.
(A) Leukemia clonal diversity across different tissues and organs from one representative mouse.
Each color represents one distinct genetic barcode corresponding to a leukemia clone. The sizes
of each colored columns indicate their relative abundance. (B) Clonal abundance comparison
between the blood and the spleen. (C) Comparison of clonal abundance between the blood and the
BM from legs. Markers of the same shape represent data from the same mouse. 99% confidence
intervals were determined by the blood and spleen comparison and highlighted by dashed lines.
Markers with black outlines depict the clones that were matched to single-cell RNA sequencing
data and used for the analysis in Fig. 3.4.

50

Figure 3.3. Bridging barcode clonal behavior with single-cell RNA sequencing data.
Human B-ALL cells are genetically labeled with DNA barcodes and transplanted into irradiated
immunocompromised mice. At the end point, barcoded cells were sorted from the blood, spleen,
bone marrow and sites of extramedullary expansion. Clonal abundance was assessed based on
barcode analysis and mapped to the single-cell RNA sequencing data as the genetic tracking
barcodes were transcribed.

51

Figure 3.4. Differentially expressed genes between blood and bone marrow.
(A) Genes significantly differentially expressed between the blood and bone marrow from legs.
The black bar indicates the mean, and the white dot represents the median. (B) All genes tested in
comparing clones that are over and under-represented in the bone marrow compared to the blood.
Shown are False Positive score (FPS) and P values of differential expression. Each dot represents
a gene. Dashed lines are at 0.05 and 0.05 for FPS and P value. FPS calculation is described in the
method section.

52

Figure 3.5. Extramedullary expansion in patient sample JFK93 naïve.
(A) Representative images showing the extramedullary site of leukemia expansion. (B) Clonal
distribution across different tissues and organs in representative mice. Each color represents one
distinct genetic barcode corresponding to a leukemia clone. The sizes of each colored columns
indicate their relative abundance. (C) Clonal abundance comparison between the blood and the
spleen. (D) Comparison of clonal abundance between the blood and the site of extramedullary
expansion. Markers of the same shape represent data from the same mouse. 99% confidence
intervals were determined by the blood and spleen comparison and highlighted by dashed lines.

53

Figure 3.6. Extramedullary expansion in patient sample JFK88 naïve.
(A) Representative images showing the extramedullary sites of leukemia expansion. (B) Clonal
distribution across different tissues and organs in representative mice. Each color represents one
distinct genetic barcode corresponding to a leukemia clone. The sizes of each colored columns
indicate their relative abundance. (C) Clonal abundance comparison between the blood and the
spleen. (D) Comparison of clonal abundance between the blood and the site of extramedullary
expansion. Markers of the same shape represent data from the same mouse. 99% confidence
intervals were determined by the blood and spleen comparison and highlighted by dashed lines.

54

Figure 3.7. Extramedullary expansion in patient sample ALL04.
(A) Representative images showing the extramedullary sites of leukemia expansion. (B) Clonal
distribution across different tissues and organs in representative mice. Each color represents one
distinct genetic barcode corresponding to a leukemia clone. The sizes of each colored columns
indicate their relative abundance. (C) Clonal abundance comparison between the blood and the
spleen. (D) Comparison of clonal abundance between the blood and the site of extramedullary
expansion. Markers of the same shape represent data from the same mouse. 99% confidence
intervals were determined by the blood and spleen comparison and highlighted by dashed lines.
Markers with black outlines depict the clones that were matched to single-cell RNA sequencing
data and used for the analysis in Fig. 3.8.

55

Figure 3.8. Differentially expressed genes between blood and ovary.
(A) The CMC2 gene was significantly upregulated in clones more abundant in the ovary than
clones more abundant in the blood. The black bar indicates the mean, and the white dot represents
the median. (B) All genes tested in comparing clones that are over and under-represented in ovary.
Shown are false positive Score (FPS) and P values of differential expression. Each dot represents
a gene. Dashed lines are at 0.05 and 0.05 for FPS and P value. FPS calculation is described in the
method section.

56
Table 3.1. Pearson correlation values from the comparison of blood, spleen and bone marrow
from spine.

57
Table 3.2. Pearson correlation values from the comparison of bone marrow from different
anatomical sites.

58
3.6 Supplementary figures

Supplemental figure 3.1
59
Supplemental Figure 3.1. Clonal distribution across different tissues and organs from all
primary mice.
Each color represents one distinct genetic barcode corresponding to a leukemia clone. The sizes
of each colored columns indicate their relative abundance. Red arrows highlighted the leg bone
marrow that exhibited different clonal compositions.

60

Supplemental figure 3.2
61
Supplemental Figure 3.2. Clonal distribution across different tissues and organs from
secondary recipient mice.
Secondary recipient mice transplanted with leukemia cells from patient (A) JFK93 naïve and (B)
ALL04. Each color represents one distinct genetic barcode corresponding to a leukemia clone. The
sizes of each colored columns indicate their relative abundance. Red arrows highlighted the leg
bone marrow that exhibited different clonal compositions.

62

Supplemental Figure 3.3. Images of all mice that were xenografted with the human B-ALL
samples.
(A) Patient ALL04, (B) JFK93 naïve, (C) JKF88 relapsed, and (D) JFK88 naïve. Significant
extramedullary expansion was consistently observed in all experimental mice. Quarter is shown
as size reference.

63

Supplemental Figure 3.4. Clonal abundance across different tissues and organs of mice that
were xenografted with patient sample ALL04.
Significant extramedullary expansion in the ovaries were observed in all replicate mice. Each color
represents one distinct genetic barcode corresponding to a leukemia clone. The sizes of each
colored columns indicate their relative abundance.

64
4 CHAPTER 4

ACUTE LYMPHOBLASTIC LEUKEMIA CLONES DIFFERENTIALLY
RESPOND TO INTENSIVE AND MAINTENANCE THERAPY

4.1 Abstract
Cellular heterogeneity is a cause of treatment resistance in cancer, however its underlying
mechanisms are largely unknown. The treatment of acute lymphoblastic leukemia (ALL) with
chemotherapy is one of the great success stories in medical oncology which has transformed a
universally fatal disease to one where most children and many adults can be cured. However, it
involves a complex treatment regimen of an intensive and maintenance phase which was derived
empirically, from decades of methodical clinical research. A mechanistic understanding of how
intensive and maintenance therapies work together has not been presented. We hypothesized, that
intensive and maintenance therapies target different ALL clone populations. To address this, we
combined in vivo single-cell tracking with single-cell mRNA sequencing to study cellular
heterogeneity in a patient-derived xenograft (PDX) model of B-cell ALL (B-ALL). We find that
small subsets of ALL clones with distinct gene expression signatures responded differently to
intensive and maintenance chemotherapies.

65
4.2 Introduction
Acute lymphoblastic leukemia
Acute lymphoblastic leukemia (ALL) is a hematologic disease characterized by the
uncontrolled proliferation of lymphoid progenitor cells in the bone marrow, blood and
extramedullary sites. The annual incidence is 1.7 cases per 100,000 people in the USA (80). In
2019, it is estimated that there will be a total of 5,930 new cases of ALL and 1,500 people will of
the disease (80). In the past decade, there have been significant advances toward understanding
the disease pathogenesis as well as advances in the development of targeted therapies aimed at the
distinct subsets of ALL. As a result, cure rates and survival outcomes for patients have improved
drastically, primary among children However, these improvements have not translated to adult
patients, as prognosis remains poor. Only 30-40% of adult patients with ALL will achieve long-
term remission (81, 82).

Diagnosis and immunophenotype
The diagnosis of ALL generally requires that a patient present more than 20% or greater
bone marrow lymphoblasts. ALL is broadly classified into three groups which include precursor
B-cell ALL, mature B-cell ALL and T-cell ALL [NCCN clinical practice guidelines: Acute
Lymphoblastic Leukemia Version 2.2019]. Furthermore, within the B-cell lineage, the disease is
subcategorized into three groups based on the stage of B-cell maturation. The three subgroups are:
(i) early precursor B-cell (early pre-B-cell), (ii) pre-B cell and (iii) mature B-cell. Early pre-B-cell
is characterized by the presence of terminal deoxynucleotide transferase (TdT), the expression of
CD19/CD22/CD79a, and most notably the absence of CD10. Pre-B-ALL is characterized by the
66
presence of cytoplasmic immunoglobulins and CD10/CD19/CD22/CD79a expression [NCCN].
Mature B-cell is characterized by showing positive for surface immunoglobulins and clonal
lambda and kappa light chains and most notably it is negative for TdT.

Treatment overview of acute lymphoblastic leukemia
Acute lymphoblastic leukemia (ALL) is a ‘poster child’ of chemotherapy cancer treatment.
It represents a great success story in medical oncology. However, it requires the administration of
multiple cycles of combination chemotherapy which are administered in varied intensities and
under precise schedules. According to the NCCN clinical practice guidelines in oncology for ALL,
the treatment approach for ALL is one of the most complex and intensive therapies in cancer
treatment. The standard of care involves a combination of drugs with varying mechanism of action
administered at different doses and often in complex schedules. The specific treatment regimens,
which include the drug selection, dose schedule and duration differ between adults and children,
and among different subtypes of ALL [NCCN clinical practice guidelines: Acute Lymphoblastic
Leukemia Version 2.2019; Pediatric Acute Lymphoblastic Leukemia Version 1.2020]. However,
the backbone and basic treatment principal is similar, and it is derived from the multiagent
chemotherapy regimen originally developed by the Berlin-Franfurt-Münster (BFM) Group for
pediatric patients.
The treatment for ALL is typically divided into three phases: (i) the induction phase, (ii)
the consolidation phase, and (iii) the maintenance phase. Interestingly, this intricate treatment
approach has evolved empirically. This dates back to 1948, when Farber et al. described temporary
remission induced by aminopterin, a folic acid antagonist, in five children with acute leukemia
67
(83). Then in 1961 Frei et al. reported complete remission rate in 59% and a 2-year survival rate
of approximately 20% in 39 pediatric patients, using a combination of mercaptopurine and
methotrexate (84).

Induction and consolidation phase
The goal of induction therapy is to reduce tumor burden by clearing as many leukemia cells
as possible from the bone marrow to result in normalization of the blood counts, and thus achieving
a minimal residual disease. Most induction regimens typically combine four or five drugs. It
combines vincristine, anthracyclines (e.g. daunorubucin, doxorubicin), and corticosteroids (e.g.
prednisone, dexamethasone) with or without L-asparaginase and/or cyclophosphamide [NCCN].
Dexamethasone has been reported to reduce central nervous system involvement and improved
patient outcomes compared to prednisone (52). The advantage of dexamethasone has been
attributed to the better penetration of the drug into the CNS (52). In the consolidation phase, the
aim is to eliminate any ‘left over’ leukemia cells following the induction phase. The combination
and duration of treatment vary among patients but typically it consists of similar drugs used during
the induction phase.

Maintenance phase
The ultimate goal of the maintenance therapy phase is to prevent relapse of the disease
following induction and consolidation therapy. Maintenance regimens are based on a backbone of
daily 6-mercaptopurine and weekly methotrexate for a duration of 2 to 3 years (52). Patients with
B-ALL, except those with mature B-cell ALL, highly benefit from a prolonged maintenance
68
therapy. During this treatment, low doses of chemotherapy is continuously applied for two or more
years. The reason for the efficacy is poorly understood, and this benefit is not observed in other
types of leukemia. Why low doses of chemotherapy are successful at eradicating leukemia while
high doses of the same drugs fail remain a mystery that cannot be explained by our current
knowledge of leukemia growth and relapse.

Relapsed acute lymphoblastic leukemia
Most chemotherapy treatments attempt to eradicate the entire cancer population in a
shotgun approach. However, this strategy does not remove all cancer cells equally and often fails
to eradicate chemo-resistant cancer cells. These cancer cells may have unique properties which
make them responsible for the relapse in patients (85–87).
In ALL, relapses are generally classified and stratified according to three prognostic
factors: time of relapse, site of relapse, and immunophenotype (51). The time of relapse, relative
to the beginning and end of treatment, is the most important prognostic factor: the earlier the
relapse occurs, the worse the prognosis (51). Clinical data shows that patients who relapse ‘early,’
within a year or two following treatment, are particularly resistant to additional chemotherapy and
have a very poor prognosis (88–90). In contrast, patients that relapse ‘late,’ meaning after two
years following treatment, have a much better prognosis. Late relapse patients tend to remain
chemotherapy sensitive (88–90). In fact, many of these patients can be cured with chemotherapy
alone. It is unclear why late relapse remains chemotherapy sensitive while early relapse becomes
chemotherapy resistant.
69
Recent genomic studies have identified relapse-specific mutations in pediatric ALL (17, 91,
92). These studies provide insights into tumor heterogeneity and the evolutionary trajectory of
cancer cells from diagnosis to relapse. One study found that different groups of clones ‘rise’ and
‘fall’ during different disease stages (93), suggesting that clonal competition may play a role during
ALL development, progression and relapse. Nonetheless, it remains unclear why late relapse
patients remain chemotherapy sensitive while early relapse patients are chemotherapy resistant.
Indeed, deciphering the underlying mechanisms that lead to relapse and understanding how and
why relapses occur is a prerequisite to defining new treatment approaches.

4.3 Results
Experimental model recapitulates clinical ALL therapy in PDX
To test the hypothesis, that intensive and maintenance therapy target distinct leukemia
clonal populations. We transplanted identical B-ALL clones from the same patient derived
xenograft (PDX) mice into multiple mice that were subsequently treated with different types of
chemotherapies. We used a multi-agent therapy consisting of vincristine, dexamethasone and L-
asparaginase for the intensive therapy phase (94, 95) and a single-agent therapy consisting of
methotrexate for the maintenance phase, mimicking clinical practice to treat ALL patients.
Experimental mice were randomly assigned to five groups: (i) vehicle control; (ii) combination
therapy consisting of short-term intensive therapy followed by prolonged maintenance therapy,
designed to approximate the clinical ALL treatment; (iii) short-term intensive therapy; (iv)
prolonged intensive therapy; and (v) prolonged maintenance therapy. Maintenance therapy was
continuously applied throughout the life spans of the mice. Prolonged intensive therapy was
applied to the toleration of the mice as assessed by changes to their body weight (Fig. 4.1).
70
We tested these five treatments in 74 mice that received B-ALL cells from three patients,
As expected, mice that received chemotherapy survived longer than those that did not (Fig. 4.2A).
Mice that received the intensive and maintenance combination therapy consistently survived the
longest (Fig. 4.2A). The survival rate directly related to the fractions of human cells in the mouse
peripheral blood (Fig. 4.2B and fig. S4.1). Furthermore, in our model, intensive therapy effectively
and rapidly removed the vast majority of leukemia cells, and maintenance therapy suppressed
leukemia growth and delayed relapse (Fig. 4.2B). Mouse body weights dropped quickly in the
absence of chemotherapy due to the leukemia burden (Fig. 4.2C). The body weight drop was also
evident following intensive therapy (Fig. 4.2C), consistent with the toxicity of intensive treatment
observed in clinical practice. Taken together, these data demonstrate that our experimental model
recapitulates key aspects of the various phases of ALL treatment.

Clonal emergence under chemotherapy treatment
We compared leukemia clones before and after various treatments and noted the emergence
of previously undetected clones after treatment (Fig. 4.3-4.4, fig. S4.2-S4.4). In particular, non-
barcoded (GFP-) human cells from patient ALL06 were found in multiple mice that were
transplanted with FACS sorted GFP+ leukemia cells (Fig. 4.3B, Fig. 4.4B, and fig. S4.3). Our
quantitative polymerase chain reaction (QPCR) analysis showed that these GFP- human cells did
not carry any genetic barcode (Fig. S4.5). These cells could be the few cells that lost the barcode
construct or collected due to sorting errors. These non-barcoded human cells were not detected in
mice that received vehicle treatment of short intensive treatment, or any of the mice receiving
sorted GFP+ leukemia cells derived from the other six patient samples.
71
Some B-ALL clones responded differentially to intensive and maintenance chemotherapy
Since all recipient mice for each patient sample received identical leukemia clones and
exhibited similar clonal composition prior to treatments (Fig. 4.3), we were able to compare the
clonal abundance at the end time point between mice of different treatments. We found that
leukemia clones from the blood, spleen and bone marrow exhibited similar responses to the
different treatments (Fig. 4.5). Moreover, we identified clones that differentially responded to the
chemotherapy treatment (Fig. 4.6 and fig. S4.6-S4.8). For example, ALL04 clone 2 responded well
to combination therapy, prolonged intensive therapy, and maintenance therapy, but not to short-
term intensive therapy (Fig. 4.6A). ALL06 clone 4 only responded to combination therapy (Fig.
4.6B). ALL20 clone 1 did not respond to maintenance therapy but responded to all treatments that
involved an intensive therapy phase (i.e. combination therapy, short-term and prolonged intensive
therapy) (Fig. 4.6C). This data supports our hypothesis that leukemia clones differentially respond
to chemotherapy.

Differentially expressed genes associated with clonal response to chemotherapy
While no clone was found to respond better to intensive therapy than to combination
therapy, some clones from ALL04 and ALL06 responded significantly better to combination
therapy than to intensive therapy alone (Fig. 4.7A and fig. S4.9). We performed single-cell RNA
sequencing analysis on donor leukemia cells to determine if the clones that exhibit a distinct
chemotherapy response share a common gene expression signature prior to the treatment. In
ALL04 prior to treatment, the clones that responded better to combination therapy than to intensive
therapy expressed significantly lower levels of EBPL and significantly higher levels of MESDC
compared to all other clones (Fig. 4.7B and fig. S4.10). Additionally, some clones in ALL04
72
responded significantly better to maintenance therapy than to intensive therapy (Fig. 4.8A). These
clones expressed lower levels of CAPNS1 prior to treatments compared to all other clones (Fig.
4.8B and fig. S4.11). From another patient sample ALL20, some clones responded significantly
better to intensive therapy than to maintenance therapy (Fig. 4.9A). Prior to treatment, these clones
expressed higher levels of BTG2, CD38, GTF2A2, ICOSLG, ITGAE and ZRANB2 compared to
all other clones (Fig. 4.9B and fig. S4.12). BTG2 is a tumor suppressor in B-ALL (96) and a known
target of p53. It is upregulated during chemotherapy mediated apoptosis in cancer cells (97, 98).
Monoclonal antibodies targeting CD38 (daratumumab, isatuximab and MOR202) (99) have been
used in multiple clinical trials for hematopoietic malignancies. It has been reported that CD38 and
ITGAE (CD103) are activated in response to pentostatin (100), an antimetabolite drug that disrupts
nucleic acid synthesis like methotrexate. This is consistent with our findings that clones with
higher expression of these genes were less sensitive to methotrexate. Taken together, the data
provide original experimental evidence that different subsets of leukemia clones differentially
responded to intensive and maintenance therapies, and exhibited distinct gene expression
signatures

73
4.4 Discussion
Therapeutic resistance drives recurrences and represents a major hurdle to successful
clinical management in relapsed ALL. In this chapter, we combine in vivo single-cell tracking with
single-cell mRNA sequencing to elucidate any underlying molecular signatures present in
leukemia cells prior to treatment that may contribute to their sensitivity to different
chemotherapeutic regimens. We established an intensive and maintenance chemotherapy protocol
in a B-cell ALL PDX model and demonstrate that small subsets of leukemia clones with distinct
gene expression signatures respond differently to intensive and maintenance chemotherapies.
The use of a combination of intensive and maintenance therapy regimen resulted in
leukemia clones responding significantly better than to an intensive therapy regimen (Fig. 4.7 and
fig. S4.10). These clones expressed significantly lower levels of EBPL (Emopamil binding protein-
like) and significantly higher levels of MESDC (Mesoderm Development LRP Chaperone)
compared to other clones. EBPL and MESDC2 direct role in cancer is unknown. Extension of this
work to further characterize these candidate genes would be important to understand their role in
cancer resistance mechanisms.
Leukemia clones found to respond significantly better to maintenance therapy than to
intensive therapy (Fig. 4.8 and fig. S4.11) expressed lower levels of CAPNS1 (calpain small
subunit 1) prior to treatment. This is an interesting observation given that CAPNS1 is a member
of the calpains family of calcium-dependent cysteine proteases. It is involved in processes that are
essential for cellular functions including apoptosis, proliferation, migration, and adhesion.
Selective inhibition of calpains is known to reduce microvesicle (MV) release (101, 102) which
are secreted by cancer cells to protect themselves against intracellular stress. MVs released from
74
methotrexate (MTX) treated cells have been found to carry MTX (103). Inhibiting MV release by
knocking down CAPNS1 with siRNA sensitizes cancer cells to chemotherapy in a xenograft model
of prostate cancer (103). Our data suggests that leukemia clones that respond poorly to
maintenance therapy may be evading the cytotoxic effects of therapy via methotrexate (MTX)
efflux.
We found that some leukemia clones that responded significantly better to the intensive
than to the maintenance chemotherapies (Fig. 4.9 and fig. S4.12). These clones expressed higher
levels of ZRANB2, BTG2, CD38, GTF2A2, ITGAE and ICOSLG. CD38 and ITGAE (CD103)
are activation markers often expressed on leukemia cells and consistent with the idea that intensive
chemotherapy selectively targets the most highly proliferative tumor fraction. In addition, BTG2
has been found to exert its antitumor effect through p53-dependent Ras signaling transduction
pathway (104). Consistent with this, our data demonstrated that BTG2 gene expression is
significantly associated with DNA damage response during vincristine treatment of intensive
therapy.
Unlike targeted therapies that are designed to target specific cell populations,
chemotherapy affects cells at specific stages of the cell cycle and does not differentiate between
cancerous and healthy cells. Therefore, mechanisms of resistance to chemotherapies are broad and
varied. Our findings support the idea that resistance to chemotherapies in ALL patients are
associated with genes responsible for drug efflux (CAPNS1) and evasion of apoptosis via
expression of DNA damage associated genes (BTG2). Altogether, the data presented in this
chapter supports our hypothesis that intensive and maintenance therapies target different leukemia
clone populations during chemotherapy treatment.
75
4.5 Figures

Figure 4.1. Chemotherapy treatment groups.
Mice were randomly assigned into five groups: (i) vehicle control; (ii) combination therapy
consisting of short-term intensive therapy followed by prolonged maintenance therapy, designed
to approximate the clinical ALL treatment; (iii) short-term intensive therapy; (iv) prolonged
intensive therapy; and (v) prolonged maintenance therapy.

76

Figure 4.2. Experimental model recapitulates clinical ALL treatment.
(A) Kaplan-Meier survival plots of PDX mice under various treatments. Color bars illustrate the
duration of the treatment period for each group. (B) Human chimerism in the peripheral blood
assessed before and during chemotherapy. Shown are the mean of all experimental mice. Mice
variations are shown in fig. S4.1. (C) Mouse body weight before and during chemotherapy
treatments. Shown are the mean of all experimental mice. Error bar represents range.

77

Figure 4.3. Clonal abundance before and at end point of treatment.
Clonal abundance in mouse peripheral blood before treatments and at the end point of each mouse
for PDX (A) ALL20 (B) ALL06 (C) ALL04. Each column represents one mouse. Each color
represents one distinct genetic barcode corresponding to a leukemia clone. The sizes of each
colored columns indicate their relative abundance.

Figure 4.4. Clonal dynamics during intensive + maintenance chemotherapy treatment.
(A-C) Representative clonal dynamics during the combination of intensive and maintenance
chemotherapy. Additional mice are shown in fig. S4.2-S4.4 Each color represents one distinct
genetic barcode corresponding to a leukemia clone. The sizes of each colored columns indicate
their relative abundance.
78

Figure 4.5. Comparing chemotherapy response of the same leukemia clones in different
tissues.
(A-B) Comparing chemotherapy responses of the same clones in different tissues. Shown are the
log2 fold differences between the average clonal abundances of two chemotherapy treatments for
each tissue. Each dot represents one clone.

79

Figure 4.6. ALL clones differentially respond to chemotherapy treatments.
(A-C) Example of leukemia clones that responded differently under different chemotherapy
treatments. Each marker represents data from one tissue and one mouse. *** P < 0.001; ** P <
0.01; * P < 0.05.

80

Figure 4.7. ALL04 clones that responded significantly better to intensive + maintenance
therapy than to intensive therapy.
(A) Leukemia clones that responded better to the combination of intensive and maintenance
therapy than to intensive therapy. Each marker represents data from one tissue and one mouse. ***
P < 0.001; ** P < 0.01; * P < 0.05. (B) Genes significantly differentially expressed in the selected
clones shown on the left (blue) compared to all other clones from the same patient samples
(orange). The black bar indicates the mean, and the white dot represents the median.

81

Figure 4.8. ALL04 clones that responded significantly better to maintenance therapy than to
intensive therapy.
(A) Leukemia clones that responded better to maintenance therapy than to intensive therapy. Each
marker represents data from one tissue and one mouse. *** P < 0.001; ** P < 0.01; * P < 0.05. (B)
Gene significantly differentially expressed in the selected clones shown on the left (blue) compared
to all other clones from the same patient samples (orange). The black bar indicates the mean, and
the white dot represents the median.

82

Figure 4.9. ALL20 clones that responded significantly better to intensive therapy than to
maintenance therapy.
(A) Leukemia clones that responded better to intensive therapies therapy than to maintenance
therapy. Each marker represents data from one tissue and one mouse. *** P < 0.001; ** P < 0.01;
* P < 0.05. (B) Gene significantly differentially expressed in the selected clones shown above
(blue) compared to all other clones from the same patient samples (orange). The black bar indicates
the mean, and the white dot represents the median.

83
4.6 Supplemental figures

Supplemental Figure 4.1. Human chimerism during chemotherapy.
Peripheral blood was collected from mice to asses human chimerism during different
chemotherapy treatments. Each plot represents data collected from one patient derived xenografts.
Shown are the mean of all experimental mice. Error bar represents range.

84

Supplemental figure 4.2.
85
Supplemental Figure 4.2. Clonal dynamics in ALL20 xenografts during chemotherapy.
Shown are clonal abundances of all experimental mice during the treatments as follows. (A)
Vehicle ‘V’ (B) Intensive and maintenance ‘IM’ (C) Intensive ‘I’ (D) Prolonged intensive ‘pI’ (E)
Continuous Maintenance ‘M’. Graphs depict the clonal distribution across the peripheral blood
throughout treatment and the clonal distribution across different tissues and organs at the endpoint.
Each color represents one distinct genetic barcode corresponding to a leukemia clone. The sizes
of each colored columns indicate their relative abundance.

86

Supplemental figure 4.3.
87
Supplemental Figure 4.3. Clonal dynamics in ALL06 xenografts during chemotherapy.
Shown are clonal abundances of all experimental mice during the treatments as follows. (A)
Vehicle ‘V’ (B) Intensive and maintenance ‘IM’ (C) Intensive ‘I’ (D) Prolonged intensive ‘pI’ (E)
Continuous Maintenance ‘M’. Graphs depict the clonal distribution across the peripheral blood
throughout treatment and the clonal distribution across different tissues and organs at the endpoint.
Each color represents one distinct genetic barcode corresponding to a leukemia clone. The sizes
of each colored columns indicate their relative abundance.

88

Supplemental figure 4.4.
89
Supplemental Figure 4.4. Clonal dynamics in ALL04 xenografts during chemotherapy.
Shown are clonal abundances of all experimental mice during the treatments as follows. (A)
Vehicle ‘V’ (B) Intensive and maintenance ‘IM’ (C) Intensive ‘I’ (D) Prolonged intensive ‘pI’ (E)
Continuous Maintenance ‘M’. Graphs depict the clonal distribution across the peripheral blood
throughout treatment and the clonal distribution across different tissues and organs at the endpoint.
Each color represents one distinct genetic barcode corresponding to a leukemia clone. The sizes
of each colored columns indicate their relative abundance.

90

Supplemental Figure 4.5. QPCR analysis of GFP negative clones.
GFP negative cells that emerged during chemotherapy treatment in ALL06 samples were sorted
and genomic DNA was extracted for QPCR amplification. Shown are amplification curves from
sorted GFP negative (GFP- cells). Curves from GFP positive (GFP+) samples from different
treatment groups are included as positive controls.

91

Supplemental Figure 4.6. ALL04 clones that responded differently under different
chemotherapy treatments.
Each plot represents data from one clone and each marker represents data from one tissue and one
mouse. *** P < 0.001; ** P < 0.01; * P < 0.05.

92

Supplemental Figure 4.7. ALL06 clones that responded differently under different
chemotherapy treatments.
Each plot represents data from one clone and each marker represents data from one tissue and one
mouse. *** P < 0.001; ** P < 0.01; * P < 0.05.

93

Supplemental Figure 4.8. ALL20 clones that responded differently under different
chemotherapy treatments.
Each plot represents data from one clone and each marker represents data from one tissue and one
mouse. *** P < 0.001; ** P < 0.01; * P < 0.05.

Supplemental Figure 4.9. ALL06 clones that responded significantly better to intensive +
maintenance treatment than to intensive treatment.
Each marker represents data from one tissue and one mouse. *** P < 0.001; ** P < 0.01; * P <
0.05.
94

Supplemental Figure 4.10. All genes tested in ALL04 clones that responded significantly
better to the intensive + maintenance therapy than to the intensive therapy compared with
all other ALL04 clones.
Shown are False Positive Score (FPS) and P values of differential expression. Each dot represents
a gene. FPS calculation is described in the method section.

Supplemental Figure 4.11. All genes tested in ALL04 clones that responded better to
maintenance therapy than to intensive therapy compared with all other ALL04 clones.
Shown are False Positive Score (FPS) and P values of differential expression. Each dot represents
a gene. FPS calculation is described in the method section.

95

Supplemental Figure 4.12. All genes tested in ALL20 clones that responded better to
intensive therapy than to maintenance therapy compared with all other ALL20 clones.
Shown are False Positive Score (FPS) and P values of differential expression. Each dot represents
a gene. FPS calculation is described in the method section

96
5 APPENDIX
Materials and Methods
Human cells
Clinical specimens were obtained from adult patients carrying B-cell acute lymphoblastic
leukemia (B-ALL). All human subjects provided informed consent, and the study was approved
by the University of Southern California institutional review board. Mononucleated cells (MNC)
were isolated by density centrifugation using Ficoll Paque Plus, density 1.077 (GE Healthcare Bio-
Sciences) followed by two washes with Iscove's Modified Dulbecco's Medium (IMDM) (Thermo
Fisher Scientific) and were frozen for later uses. During recovery, frozen cells were thawed and
cultured in IMDM supplemented with 20% fetal bovine serum (FBS) (VWR Life Science
Seradigm) for 1 to 2 hours at 37° C. Cells were then stained, analyzed and sorted.

Leukemia cell culture and lentiviral transduction
Human B-ALL cells were sorted for human CD45 and CD19 (Table A.1) from
cryopreserved samples. These cells were either primary human bone marrow aspirates or passaged
mouse spleen cells. Cells were cultured in StemSpanTM Serum-Free Expansion Medium II (SFEM
II) (Stem Cell Technologies) in the presence of 20 ng/ml human FLT-3 ligand (Gibco by Life
Technologies), 20 ng/ml human Interleukin-3 (IL-3) (Gibco by Life Technologies) and 50 ng/ml
human Stem Cell Factor (SCF) (Gibco by Life Technologies). After 24 hours of pre-stimulation
under the culture condition, cells were washed and incubated for another 16 hours in the same
medium with addition of lentivirus carrying the DNA barcodes. 8 ng/µl polybrene was added into
97
the culture to facilitate viral transduction. B-ALL cells were washed three times with Dulbecco’s
Phosphate Buffered Saline (D-PBS) (Gibco by Life Technologies) prior to transplantation.

Mice
NOD.Cg-Prkdcscid Il2rgtm1Wjl (NSG, JAX stock number 05557) and NOD.Cg-
Prkdcscid Il2rgtm1Wjl Tg(CMV-IL3,CSF2,KITLG)1Eav/MloySzJ (NSG-SGM3, JAX stock number
013062) mice were obtained from the Jackson Laboratory. Mice were bred and maintained at the
Research Animal Facility of the University of Southern California. Animal procedures were
approved by the Institutional Animal Care and Use Committee.

Human B-ALL engraftment
NSG or NSG-SGM3 mice were irradiated with 150 cGy and transplanted with 100,000 to
200,000 human B-ALL cells via tail vein injection. Mice were monitored daily for evidence of
distress and were euthanized when human chimerism exceeded 90% of total MNC. At the
endpoint, mouse peripheral blood was collected via perfusion using D-PBS with 10mM
Ethylenediaminetetraacetic acid (EDTA) (Sigma-Aldrich). Spleen, bones, and tissues with
noticeable extramedullary expansion were collected. Single cell suspensions were prepared by
crushing the tissues in D-PBS with 2% FBS and filtered through a 70uM cell strainer. Cells were
analyzed by flow cytometry. 500,000 barcoded cells were sorted for barcode analysis. Additional
unsorted cells were frozen in IMDM with 20% FBS and 10% dimethylsulfoxide (DMSO) (Sigma-
Aldrich) for later uses.

98
Bioluminescent imaging
Lentiviral production of a pCDH-CMV-EF1-puro construct (Systems Bioscience)
containing a ubiquitin promoter driving the expression of a fusion protein containing the Luc2
(pgl4) luciferase gene (Promega) and the eGFP gene (Becton Dickinson) was carried out using
standard protocols. NSG mice were used for xenotransplantation studies using patient B-ALL cells
that were transduced with eGFP-Luc2 encoding lentivirus. Bioluminescent activity was visualized
in vivo after D-luciferin injection (Biosynth) on an IVIS Spectrum (Caliper Life Sciences)
instrument and quantified using Living Image 4.0 software. D-Luciferin (firefly) potassium salt
(Biosynth) solution was prepared by dissolving 1 gram in 60 milliliters of D-PBS. Mice were
injected intraperitoneally with luciferin solution (0.139 g luciferin per kilogram body weight) and
imaged. Total flux (photons per second) values were obtained from mice.

Blood sample collection and FACS analysis
Blood samples were collected into D-PBS containing 10 mM EDTA via a small transverse
cut in the tail vein. 2% dextran (Pharmacosmos) was added to help remove red blood cells. To
further eliminate red blood cells, the remaining blood cells were treated with ammonium-chloride-
potassium lysis buffer on ice for 5 minutes. After a 45-60-minute antibody incubation at 4° C,
samples were suspended in D-PBS with 2% FBS and 4,6-Diamidino-2-phenylindole to distinguish
dead cells. Cells were analyzed and sorted using the FACS-Aria I and II cell sorters. Antibodies
were obtained from eBioscience (currently Life Technologies/Thermo Fisher) and BioLegend
(Table A.1). Flow cytometry data were analyzed using Diva software 8.0.1 (BD Biosciences).

99
Chemotherapy treatment
Barcoded human B-ALL cells were transplanted into sub lethally irradiated NSG or NSG-
SGM3 mice. Once human leukemia cell contribution reached 20-40% of total MNC, mice were
randomized and placed into one of the five chemotherapy groups. The five groups were: (i) vehicle
(V) control, (ii) short-term intensive (I) therapy, (iii) short-term intensive therapy followed by
long-term maintenance (IM) therapy, (iv) prolonged intensive (pI) therapy, and (v) maintenance
(M) therapy. The chemotherapy treatment consisted of an intensive and a maintenance phase. The
intensive therapy phase consisted of Vincristine (Hospira Pharmaceuticals) (0.25mg/kg)
administered weekly via intravenous (IV) injection, Dexamethasone (AuroMedics Pharma)
(7.5mg/kg) administered Monday, Wednesday, Friday via intraperitoneal (IP) injection, and L-
asparaginase (Sigma-Tau Pharmaceuticals) (100 IU/kg) administered bi-weekly via IP injection.
Maintenance therapy phase consisted of weekly Methotrexate (Accord Healthcare) (5mg/kg)
administered via IV or intramuscular injection.
Vehicle group received weekly IV injections of Bacteriostatic Water (Hospira
Pharmaceuticals) during the life span of the mice. The short-term intensive therapy group only
received four weeks of intensive therapy. The combination therapy group received four weeks of
intensive therapy followed with maintenance therapy. The prolonged intensive therapy group was
treated with intensive therapy for as long as the mice tolerated high dose treatment. Toxicity was
assessed based on body weight change. The range of treatment was 7 weeks for ALL04 and ALL20
and 10 weeks for ALL06. The maintenance therapy group was treated with maintenance therapy
continuously during the life span of the mice. The range of treatment was 5 weeks for ALL20, 14
weeks for ALL04 and 28 weeks for ALL06.
Leukemia progression was monitored throughout the duration of treatment by analyzing
the peripheral blood. In addition, mouse weight was monitored weekly throughout treatment to
100
asses signs of therapy toxicity. If body weight dropped more than 20% of the starting weight,
chemotherapy doses were adjusted. Animal care was in accordance with institutional guidelines.
Survival (Kaplan-Meier) curves were generated by using GraphPad Prism Software (San Diego,
California) with significance determined by the log-rank test.

DNA barcode extraction and analysis
Genomic DNA was extracted from sorted barcoded leukemia cells and amplified using
Phusion PCR master mix (Thermo Scientific, Waltham, MA). The PCR reactions were halted once
they had progressed halfway through the exponential phase. PCR product was purified and
analyzed using high-throughput sequencing. We combined sequencing data with FACS data to
calculate the clonal abundance for each clone as indicated below. Clones with a clonal abundance
greater than 0.01% were used for further analyses.
𝐶𝑙𝑜𝑛𝑎𝑙 𝑎𝑏𝑢𝑛𝑑𝑎𝑛𝑐𝑒 % = 100 ∗ [
# 𝑜𝑓 𝑟𝑒𝑎𝑑𝑠 𝑓𝑜𝑟 𝑒𝑎𝑐 ℎ 𝑏𝑎𝑟𝑐𝑜𝑑𝑒 𝑡𝑜𝑡𝑎𝑙 𝑟𝑒𝑎𝑑𝑠 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑏𝑎𝑟𝑐𝑜𝑑𝑒𝑠 ] [
# 𝑜𝑓 ℎ𝑢𝑚𝑎𝑛 𝑐𝑒𝑙𝑙𝑠 𝑡𝑜𝑡𝑎𝑙 𝑀𝑁𝐶 ] [
# 𝑜𝑓 𝐺𝐹𝑃 𝑐𝑒𝑙𝑙𝑠 𝑡𝑜𝑡𝑎𝑙 ℎ𝑢𝑚𝑎𝑛 𝑐𝑒𝑙 𝑙𝑠
]

Single-cell RNA sequencing and data analysis
Single-cell RNA sequencing (scRNA-seq) was performed following the manufacturer’s
protocol for the Chromium Single Cell 3’ Library (10X Genomics, V2) with minor modifications
as following. After cDNA amplification, half of the amplified cDNA was used for the downstream
fragmentation, adaptor ligation and sample index PCR. The other half of the amplified cDNA was
used for extracting and amplifying molecules that contain both the genetic tracking barcodes and
the Chromium cellular barcodes. The cDNA libraries were sequenced using an Illumina HiSeq
2500 at a coverage of 50,000 raw reads per cell (paired end; read1: 26 cycles; i7 index: 8 cycles;
101
read 2: 98 cycles). Raw data were processed using the Cell Ranger pipeline (10X Genomics, v
2.1.0) for cellular barcode assignment and unique molecule identifier (UMI) quantification. Cells
with more than 10% UMIs mapped to mitochondrial genes were excluded. Genes with more than
2 UMIs in more than 0.5% of cells were used for downstream analyses. Expression values for gene
i in cell j were calculated by dividing UMI count values for gene i by the sum of the UMI counts
in cell j, and then multiplying by 10,000 to create transcripts per million (TPM)-like values, and
finally calculating log
2
(𝑇𝑃𝑀 + 1) as gene expression values. For comparing single-cell gene
expression data, P values were calculated using the one-sided Mann Whitney U-test.

False positive score calculation
False Positive Score (FPS) was calculated by comparing experimental data and scramble
data. Five sets of scramble data were generated by randomly mapping tracking barcode data to
gene expression data. For each gene, P values were calculated by the one-sided Mann Whitney U-
test using both the experimental data and the scramble data. FPS for each P value of the
experimental data was then calculated as the number of genes whose P values were equal or smaller
than this P value in the scramble data (median of the five sets) divided by that gene number in the
experimental data. Genes with FPS<0.05 and P value <0.05 were considered significant.

Statistical analysis
Diversity of clones was calculated using the Shannon Diversity Index, as implemented by
python package skbio.diversity (scikit-bio). K-Clustering method (Fig. 2.6) was performed on
filtered clonal abundance data (clones that contributed more than 0.01% of MNC) as implemented
by python package sklearn.cluster.Kmeans (scikit-learn), the number of clusters was set to 2,
102
which was determined by the smallest inertia and all other parameters are default. Tissue biases
(chapter 3 figures) were determined based on clonal abundance variations between blood and
spleen (99% confidence interval), assuming non-biased clones have equal abundance across
tissues. P-values for clonal response across treatment groups were calculated via independent t-
test with equal variance assumed. Significance in all figures was indicated as follows: ***P<0.001;
**P<0.01; *P<0.05; n.s. P>0.05.

Table A.1. Monoclonal antibodies used in this study.
Antigen Conjugate Vendor Catalog # Clone Lot #
mouse CD45 Alexa 700 Biolegend 110724 A20 B254605
human CD45 APC-eFluor 780 ebioscience 47-0459-42 H130 4331639
human CD19 PE ebioscience 12-0198-42 SJ25C1 1987712

103
References

1. R. Sun, Z. Hu, C. Curtis, Big Bang Tumor Growth and Clonal Evolution, Csh Perspect Med 8,
a028381 (2018).

2. I. U. Group, G. Gundem, P. Loo, B. Kremeyer, L. exandrov, J. M. Tubio, E. Papaemmanuil,
D. S. Brewer, H. M. Kallio, G. Högnäs, M. Annala, K. Kivinummi, V. Goody, C. Latimer, S.
O’Meara, K. J. Dawson, W. Isaacs, M. R. Emmert-Buck, M. Nykter, C. Foster, Z. Kote-Jarai, D.
Easton, H. C. Whitaker, D. E. Neal, C. S. Cooper, R. A. Eeles, T. Visakorpi, P. J. Campbell, U.
rmott, D. C. Wedge, S. G. Bova, The evolutionary history of lethal metastatic prostate cancer,
Nature 520, 353–357 (2015).

3. L. R. Yates, M. Gerstung, S. Knappskog, C. Desmedt, G. Gundem, P. Loo, T. Aas, L.
exandrov, D. Larsimont, H. Davies, Y. Li, Y. Ju, M. Ramakrishna, H. Haugland, P. Lilleng, S.
Nik-Zainal, S. McLaren, A. Butler, S. Martin, D. Glodzik, A. Menzies, K. Raine, J. Hinton, D.
Jones, L. J. Mudie, B. Jiang, D. Vincent, A. Greene-Colozzi, P.-Y. Adnet, A. Fatima, M.
Maetens, M. Ignatiadis, M. R. Stratton, C. Sotiriou, A. L. Richardson, P. Lønning, D. C. Wedge,
P. J. Campbell, Subclonal diversification of primary breast cancer revealed by multiregion
sequencing, Nat Med 21, 751–759 (2015).

4. A. L. Paguirigan, J. Smith, S. Meshinchi, M. Carroll, C. Maley, J. P. Radich, Single-cell
genotyping demonstrates complex clonal diversity in acute myeloid leukemia., Science
translational medicine 7, 281re2 (2015).

5. L. Melchor, A. Brioli, C. Wardell, A. Murison, N. Potter, Single-cell genetic analysis reveals
the composition of initiating clones and phylogenetic patterns of branching and parallel evolution
in myeloma, (2014).

6. R. Lu, N. F. Neff, S. R. Quake, I. L. Weissman, Tracking single hematopoietic stem cells in
vivo using high-throughput sequencing in conjunction with viral genetic barcoding., Nature
biotechnology 29, 928–33 (2011).
104

7. J. S. Welch, T. J. Ley, D. C. Link, C. A. Miller, D. E. Larson, D. C. Koboldt, L. D. Wartman,
T. L. Lamprecht, F. Liu, J. Xia, C. Kandoth, R. S. Fulton, M. D. McLellan, D. J. Dooling, J. W.
Wallis, K. Chen, C. C. Harris, H. K. Schmidt, J. M. Kalicki-Veizer, C. Lu, Q. Zhang, L. Lin, M.
D. O’Laughlin, J. F. McMichael, K. D. Delehaunty, L. A. Fulton, V. J. Magrini, S. D. McGrath,
R. T. Demeter, T. L. Vickery, J. Hundal, L. L. Cook, G. W. Swift, J. P. Reed, P. A. Alldredge, T.
N. Wylie, J. R. Walker, M. A. Watson, S. E. Heath, W. D. Shannon, N. Varghese, R. Nagarajan,
J. E. Payton, J. D. Baty, S. Kulkarni, J. M. Klco, M. H. Tomasson, P. Westervelt, M. J. Walter,
T. A. Graubert, J. F. DiPersio, L. Ding, E. R. Mardis, R. K. Wilson, The Origin and Evolution of
Mutations in Acute Myeloid Leukemia, Cell 150, 264–278 (2012).

8. N. Coccaro, L. Anelli, A. Zagaria, G. Specchia, F. Albano, Next-Generation Sequencing in
Acute Lymphoblastic Leukemia, Int J Mol Sci 20, 2929 (2019).

9. B. M. Faltas, D. Prandi, S. T. Tagawa, A. M. Molina, D. M. Nanus, C. Sternberg, J.
Rosenberg, J. Mosquera, B. Robinson, O. Elemento, A. Sboner, H. Beltran, F. Demichelis, M. A.
Rubin, Clonal evolution of chemotherapy-resistant urothelial carcinoma, Nat Genet 48, 1490–
1499 (2016).

10. J. Saliba, N. Evensen, J. Meyer, I. Dolgalev, D. Newman, A. Chowdhury, J. Nersting, J.
Wang, K. Schmiegelow, W. L. Carroll, Using Whole Exome Sequencing in Pediatric Acute
Lymphoblastic Leukemia Germline, Diagnosis, and Relapse Trios to Discover Novel Relapse
Enriched Mutations for Clonal Backtracking By Ddpcr, Blood 128, 4085–4085 (2016).

11. H. Lilljebjörn, M. Rissler, C. Lassen, J. Heldrup, M. Behrendtz, F. Mitelman, B. Johansson,
T. Fioretos, Whole-exome sequencing of pediatric acute lymphoblastic leukemia, Leukemia 26,
1602–1607 (2012).

12. A. Roth, J. Khattra, D. Yap, A. Wan, E. Laks, J. Biele, G. Ha, S. Aparicio, A. Bouchard-
Côté, S. P. Shah, PyClone: statistical inference of clonal population structure in cancer, Nat
Methods 11, 396–398 (2014).
105

13. C. A. Miller, B. S. White, N. D. Dees, M. Griffith, J. S. Welch, O. L. Griffith, R. Vij, M. H.
Tomasson, T. A. Graubert, M. J. Walter, M. J. Ellis, W. Schierding, J. F. DiPersio, T. J. Ley, E.
R. Mardis, R. K. Wilson, L. Ding, SciClone: Inferring Clonal Architecture and Tracking the
Spatial and Temporal Patterns of Tumor Evolution, Plos Comput Biol 10, e1003665 (2014).

14. A. G. Deshwar, S. Vembu, C. K. Yung, G. Jang, L. Stein, Q. Morris, PhyloWGS:
Reconstructing subclonal composition and evolution from whole-genome sequencing of tumors,
Genome Biol 16, 35 (2015).

15. C. G. Mullighan, L. A. Phillips, X. Su, J. Ma, C. B. Miller, S. A. Shurtleff, J. R. Downing,
Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia., Science (New
York, N.Y.) 322, 1377–80 (2008).

16. C. le Viseur, M. Hotfilder, S. Bomken, K. Wilson, S. Röttgers, A. Schrauder, A. Rosemann,
J. Irving, R. W. Stam, L. D. Shultz, J. Harbott, H. Jürgens, M. Schrappe, R. Pieters, J. Vormoor,
In childhood acute lymphoblastic leukemia, blasts at different stages of immunophenotypic
maturation have stem cell properties., Cancer Cell 14, 47–58 (2008).

17. C. G. Mullighan, J. Zhang, L. H. Kasper, S. Lerach, D. Payne-Turner, L. A. Phillips, S. L.
Heatley, L. Holmfeldt, R. J. Collins-Underwood, J. Ma, CREBBP mutations in relapsed acute
lymphoblastic leukaemia, Nature 471, 235 (2011).

18. Z. Gu, M. L. Churchman, K. G. Roberts, I. Moore, X. Zhou, J. Nakitandwe, K. Hagiwara, S.
Pelletier, S. Gingras, H. Berns, D. Payne-Turner, A. Hill, I. Iacobucci, L. Shi, S. Pounds, C.
Cheng, D. Pei, C. Qu, S. Newman, M. Devidas, Y. Dai, S. C. Reshmi, J. Gastier-Foster, E. A.
Raetz, M. J. Borowitz, B. L. Wood, W. L. Carroll, P. A. Zweidler-McKay, K. R. Rabin, L. A.
Mattano, K. W. Maloney, A. Rambaldi, O. Spinelli, J. P. Radich, M. D. Minden, J. M. Rowe, S.
Luger, M. R. Litzow, M. S. Tallman, J. Racevskis, Y. Zhang, R. Bhatia, J. Kohlschmidt, K.
Mrózek, C. D. Bloomfield, W. Stock, S. Kornblau, H. M. Kantarjian, M. Konopleva, W. E.
Evans, S. Jeha, C.-H. H. Pui, J. Yang, E. Paietta, J. R. Downing, M. V. Relling, J. Zhang, M. L.
106
Loh, S. P. Hunger, C. G. Mullighan, PAX5-driven subtypes of B-progenitor acute lymphoblastic
leukemia., Nature genetics 51, 296–307 (2019).

19. C. G. Mullighan, Genomic Characterization of Childhood Acute Lymphoblastic Leukemia,
Semin Hematol 50, 314–324 (2013).

20. T. Yasuda, D. Nishijima, S. Kojima, M. Kawazu, T. Ueno, S. Tsuzuki, H. Kiyoi, I.
Matsumura, Y. Miyazaki, K. Horibe, H. Mano, T. Naoe, M. Sanada, F. Hayakawa, Genomic and
Clinical Characterization of Adult Ph-Negative B-Cell Acute Lymphoblastic Leukemia, Blood
132, 2821–2821 (2018).

21. C. Gawad, W. Koh, S. R. Quake, Dissecting the clonal origins of childhood acute
lymphoblastic leukemia by single-cell genomics., Proceedings of the National Academy of
Sciences of the United States of America 111, 17947–52 (2014).

22. C. Yu, J. Yu, X. Yao, W. K. Wu, Y. Lu, S. Tang, X. Li, L. Bao, X. Li, Y. Hou, R. Wu, M.
Jian, R. Chen, F. Zhang, L. Xu, F. Fan, J. He, Q. Liang, H. Wang, X. Hu, M. He, X. Zhang, H.
Zheng, Q. Li, H. Wu, Y. Chen, X. Yang, S. Zhu, X. Xu, H. Yang, J. Wang, X. Zhang, J. J. Sung,
Y. Li, J. Wang, Discovery of biclonal origin and a novel oncogene SLC12A5 in colon cancer by
single-cell sequencing, Cell Res 24, 701–712 (2014).

23. C. Kim, R. Gao, E. Sei, R. Brandt, J. Hartman, T. Hatschek, N. Crosetto, T. Foukakis, N. E.
Navin, Chemoresistance Evolution in Triple-Negative Breast Cancer Delineated by Single-Cell
Sequencing, Cell 173, 879-893.e13 (2018).

24. Y. J. Tang, J. Huang, H. Tsushima, G. I. Ban, H. Zhang, K. M. Oristian, V. Puviindran, N.
Williams, X. Ding, J. Ou, S.-H. Jung, C.-L. Lee, Y. Jiao, B. J. Chen, D. G. Kirsch, B. A. Alman,
Tracing Tumor Evolution in Sarcoma Reveals Clonal Origin of Advanced Metastasis, Cell
Reports 28, 2837-2850.e5 (2019).

25. L. V. Nguyen, C. L. Cox, P. Eirew, D. J. Knapp, D. Pellacani, N. Kannan, A. Carles, M.
107
Moksa, S. Balani, S. Shah, M. Hirst, S. Aparicio, C. J. Eaves, DNA barcoding reveals diverse
growth kinetics of human breast tumour subclones in serially passaged xenografts., Nature
communications 5, 5871 (2014).

26. L. V. Nguyen, D. Pellacani, S. Lefort, N. Kannan, T. Osako, M. Makarem, C. L. Cox, W.
Kennedy, P. Beer, A. Carles, M. Moksa, M. Bilenky, S. Balani, S. Babovic, I. Sun, M. Rosin, S.
Aparicio, M. Hirst, C. J. Eaves, Barcoding reveals complex clonal dynamics of de novo
transformed human mammary cells., Nature 528, 267–71 (2015).

27. C. Hyo-eun, D. Ruddy, V. Radhakrishna, Studying clonal dynamics in response to cancer
therapy using high-complexity barcoding, Nature Medicine (2015), doi:10.1038/nm.3841 .

28. A. N. Hata, M. J. Niederst, H. L. Archibald, M. Gomez-Caraballo, F. ddiqui, H. E. Mulvey,
Y. E. Maruvka, F. Ji, H. C. Bhang, V. Radhakrishna, G. Siravegna, H. Hu, S. Raoof, E.
Lockerman, A. Kalsy, D. Lee, C. L. Keating, D. A. Ruddy, L. J. Damon, A. S. Crystal, C. Costa,
Z. Piotrowska, A. Bardelli, A. J. Iafrate, R. I. Sadreyev, F. Stegmeier, G. Getz, L. V. Sequist, A.
C. Faber, J. A. Engelman, Tumor cells can follow distinct evolutionary paths to become resistant
to epidermal growth factor receptor inhibition, Nat Med 22, 262–269 (2016).

29. D. Merino, T. Weber, A. Serrano, F. Vaillant, K. Liu, B. Pal, D. L. Stefano, J. Schreuder, D.
Lin, Y. Chen, M. Asselin-Labat, T. Schumacher, D. Cameron, G. Smyth, A. Papenfuss, G.
Lindeman, J. Visvader, S. Naik, Barcoding reveals complex clonal behavior in patient-derived
xenografts of metastatic triple negative breast cancer, Nat Commun 10, 766 (2019).

30. C. Brewer, E. Chu, M. Chin, R. Lu, Transplantation Dose Alters the Differentiation Program
of Hematopoietic Stem Cells., Cell reports 15, 1848–57 (2016).

31. L. Nguyen, Z. Wang, A. Y. Chowdhury, E. Chu, J. Eerdeng, D. Jiang, R. Lu, Functional
compensation between hematopoietic stem cell clones in vivo., EMBO reports 19 (2018),
doi:10.15252/embr.201745702 .

108
32. R. Lu, A. Czechowicz, J. Seita, D. Jiang, I. L. Weissman, Clonal-level lineage commitment
pathways of hematopoietic stem cells in vivo., Proceedings of the National Academy of Sciences
of the United States of America 116, 1447–1456 (2019).

33. L. Mascarenhas, R. Stripecke, S. S. Case, D. Xu, K. I. Weinberg, D. B. Kohn, Gene Delivery
to Human B-Precursor Acute Lymphoblastic Leukemia Cells, Blood 92, 3537–3545 (1998).

34. R. Stripecke, A. A. Cardoso, K. A. Pepper, D. C. Skelton, X.-J. Yu, L. Mascarenhas, K. I.
Weinberg, L. M. Nadler, D. B. Kohn, Lentiviral vectors for efficient delivery of CD80 and
granulocyte-macrophage– colony-stimulating factor in human acute lymphoblastic leukemia and
acute myeloid leukemia cells to induce antileukemic immune responses, Blood 96, 1317–1326
(2000).

35. E. Biagi, F. Bambacioni, G. Gaipa, C. Casati, J. Golay, A. Biondi, M. Introna, Efficient
lentiviral transduction of primary human acute myelogenous and lymphoblastic leukemia cells.,
Haematologica 86, 13–6 (2001).

36. S. B. Willingham, J.-P. P. Volkmer, A. J. Gentles, D. Sahoo, P. Dalerba, S. S. Mitra, J.
Wang, H. Contreras-Trujillo, R. Martin, J. D. Cohen, P. Lovelace, F. A. Scheeren, M. P. Chao,
K. Weiskopf, C. Tang, A. K. Volkmer, T. J. Naik, T. A. Storm, A. R. Mosley, B. Edris, S. hmid,
C. K. Sun, M.-S. S. Chua, O. Murillo, P. Rajendran, A. C. Cha, R. K. Chin, D. Kim, M. Adorno,
T. Raveh, D. Tseng, S. Jaiswal, P. Enger, G. K. Steinberg, G. Li, S. K. So, R. Majeti, G. R.
Harsh, M. van de Rijn, N. N. Teng, J. B. Sunwoo, A. A. Alizadeh, M. F. Clarke, I. L. Weissman,
The CD47-signal regulatory protein alpha (SIRPa) interaction is a therapeutic target for human
solid tumors., Proceedings of the National Academy of Sciences of the United States of America
109, 6662–7 (2012).

37. B. Edris, S. B. Willingham, K. Weiskopf, A. K. Volkmer, J.-P. P. Volkmer, T. Mühlenberg,
K. D. Montgomery, H. Contreras-Trujillo, A. Czechowicz, J. A. Fletcher, R. B. West, I. L.
Weissman, M. van de Rijn, Anti-KIT monoclonal antibody inhibits imatinib-resistant
gastrointestinal stromal tumor growth., Proceedings of the National Academy of Sciences of the
109
United States of America 110, 3501–6 (2013).

38. S. Li, D. Shen, J. Shao, R. Crowder, W. Liu, A. Prat, X. He, S. Liu, J. Hoog, C. Lu, L. Ding,
O. L. Griffith, C. Miller, D. Larson, R. S. Fulton, M. Harrison, T. Mooney, J. F. McMichael, J.
Luo, Y. Tao, R. Goncalves, C. Schlosberg, J. F. Hiken, L. Saied, C. Sanchez, T. Giuntoli, C.
Bumb, C. Cooper, R. T. Kitchens, A. Lin, C. Phommaly, S. R. Davies, J. Zhang, M. Kavuri, D.
McEachern, Y. Dong, C. Ma, T. Pluard, M. Naughton, R. Bose, R. Suresh, R. well, L. Michel, R.
Aft, W. Gillanders, K. DeSchryver, R. K. Wilson, S. Wang, G. B. Mills, A. Gonzalez-Angulo, J.
R. Edwards, C. Maher, C. M. Perou, E. R. Mardis, M. J. Ellis, Endocrine-Therapy-Resistant
ESR1 Variants Revealed by Genomic Characterization of Breast-Cancer-Derived Xenografts,
Cell Reports 4, 1116–1130 (2013).

39. U. Ben-David, G. Ha, Y.-Y. Tseng, N. F. Greenwald, C. Oh, J. Shih, J. M. McFarland, B.
Wong, J. S. Boehm, R. Beroukhim, T. R. Golub, Patient-derived xenografts undergo mouse-
specific tumor evolution, Nat Genet 49, 1567–1575 (2017).

40. S. H. Kresse, L. A. Meza-Zepeda, I. Machado, A. Llombart-Bosch, O. Myklebost, Preclinical
xenograft models of human sarcoma show nonrandom loss of aberrations., Cancer 118, 558–70
(2011).

41. L. Ding, M. J. Ellis, S. Li, D. E. Larson, K. Chen, J. W. Wallis, C. C. Harris, M. D.
McLellan, R. S. Fulton, L. L. Fulton, R. M. Abbott, J. Hoog, D. J. Dooling, D. C. Koboldt, H.
Schmidt, J. Kalicki, Q. Zhang, L. Chen, L. Lin, M. C. Wendl, J. F. McMichael, V. J. Magrini, L.
Cook, S. D. McGrath, T. L. Vickery, E. Appelbaum, K. Deschryver, S. Davies, T. Guintoli, L.
Lin, R. Crowder, Y. Tao, J. E. Snider, S. ith, A. F. Dukes, G. E. Sanderson, C. S. Pohl, K. D.
Delehaunty, C. C. Fronick, K. A. Pape, J. S. Reed, J. S. Robinson, J. S. Hodges, W. Schierding,
N. D. Dees, D. Shen, D. P. Locke, M. E. Wiechert, J. M. Eldred, J. B. Peck, B. J. Oberkfell, J. T.
Lolofie, F. Du, A. E. Hawkins, M. D. O’Laughlin, K. E. Bernard, M. Cunningham, G. Elliott, M.
D. Mason, D. M. Thompson, J. L. Ivanovich, P. J. Goodfellow, C. M. Perou, G. M. Weinstock,
R. Aft, M. Watson, T. J. Ley, R. K. Wilson, E. R. Mardis, Genome remodelling in a basal-like
breast cancer metastasis and xenograft., Nature 464, 999–1005 (2010).
110

42. A. Kreso, C. A. O’Brien, P. van Galen, O. I. Gan, F. Notta, A. M. Brown, K. Ng, J. Ma, E.
Wienholds, C. Dunant, A. Pollett, S. Gallinger, J. McPherson, C. G. Mullighan, D. Shibata, J. E.
Dick, Variable clonal repopulation dynamics influence chemotherapy response in colorectal
cancer., Science (New York, N.Y.) 339, 543–8 (2013).

43. A. Baldi, T. Battista, A. Luca, D. Santini, L. Rossiello, F. Baldi, P. Natali, D. Lombardi, M.
Picardo, A. Felsani, M. G. Paggi, Identification of genes down‐regulated during melanoma
progression: a cDNA array study, Exp Dermatol 12, 213–218 (2003).

44. S.-M. Park, H. Cho, A. M. Thornton, T. S. Barlowe, T. Chou, S. Chhangawala, L. Fairchild,
J. Taggart, A. Chow, A. Schurer, A. Gruet, M. D. Witkin, J. Kim, E. evach, A. Krivtsov, S. A.
Armstrong, C. Leslie, M. G. Kharas, IKZF2 Drives Leukemia Stem Cell Self-Renewal and
Inhibits Myeloid Differentiation, Cell Stem Cell 24, 153-165.e7 (2019).

45. R. Zappasodi, G. Ruggiero, C. Guarnotta, M. Tortoreto, C. Tringali, A. Cavanè, A. D.
Cabras, L. Castagnoli, B. Venerando, N. Zaffaroni, A. M. Gianni, F. Braud, C. Tripodo, S. M.
Pupa, M. Nicola, HSPH1 inhibition downregulates Bcl-6 and c-Myc and hampers the growth of
human aggressive B-cell non-Hodgkin lymphoma, Blood 125, 1768–1771 (2015).

46. C. F. Wright, B. W. Oswald, S. Dellis, Vaccinia Virus Late Transcription Is Activated in
Vitro by Cellular Heterogeneous Nuclear Ribonucleoproteins, J Biol Chem 276, 40680–40686
(2001).

47. S. reban, S. Ramalingam, G. Natarajan, R. May, D. Subramaniam, K. Bishnupuri, A.
Morrison, B. Dieckgraefe, D. Brackett, R. Postier, C. Houchen, S. Anant, Translation regulatory
factor RBM3 is a proto-oncogene that prevents mitotic catastrophe, Oncogene 27, 4544–4556
(2008).

48. L. V. Sequist, B. A. Waltman, D. Dias-Santagata, S. Digumarthy, A. B. Turke, P. Fidias, K.
Bergethon, A. T. Shaw, S. Gettinger, A. K. Cosper, S. Akhavanfard, R. S. Heist, J. Temel, J. G.
111
Christensen, J. C. Wain, T. J. Lynch, K. Vernovsky, E. J. Mark, M. Lanuti, J. A. Iafrate, M.
Mino-Kenudson, J. A. Engelman, Genotypic and Histological Evolution of Lung Cancers
Acquiring Resistance to EGFR Inhibitors, Sci Transl Med 3, 75ra26-75ra26 (2011).

49. J. J. Chabon, A. D. Simmons, A. F. Lovejoy, M. S. Esfahani, A. M. Newman, H. J.
Haringsma, D. M. Kurtz, H. Stehr, F. Scherer, C. A. Karlovich, T. C. Harding, K. A. Durkin, G.
A. Otterson, T. W. Purcell, R. D. Camidge, J. W. Goldman, L. V. Sequist, Z. Piotrowska, H. A.
Wakelee, J. W. Neal, A. A. Alizadeh, M. Diehn, Circulating tumour DNA profiling reveals
heterogeneity of EGFR inhibitor resistance mechanisms in lung cancer patients, Nat Commun 7,
11815 (2016).

50. I. Dagogo-Jack, A. T. Shaw, Tumour heterogeneity and resistance to cancer therapies, Nat
Rev Clin Oncol 15, 81–94 (2018).

51. D. Bhojwani, C.-H. Pui, Relapsed childhood acute lymphoblastic leukaemia., Lancet Oncol
14, e205-17 (2013).

52. J. C. Alvarnas, P. A. Brown, P. Aoun, K. Ballen, S. K. Barta, U. Borate, M. W. Boyer, P. W.
Burke, R. Cassaday, J. E. Castro, P. F. Coccia, S. E. Coutre, L. E. Damon, D. J. DeAngelo, D.
Douer, O. Frankfurt, J. P. Greer, R. A. Johnson, H. M. Kantarjian, R. B. Klisovic, G. Kupfer, M.
Litzow, A. Liu, A. V. Rao, B. Shah, G. L. Uy, E. S. Wang, A. D. Zelenetz, K. Gregory, C.
Smith, Acute Lymphoblastic Leukemia, Version 2.2015, J Natl Compr Canc Ne 13, 1240–1279
(2015).

53. E. Viadana, I. Bross, J. Pickren, An Autopsy Study of the Metastatic Patterns of Human
Leukemias, Oncology 35, 87–96 (1978).

54. R. Himelstein-Braw, H. Peters, M. Faber, Morphological study of the ovaries of leukaemic
children, Brit J Cancer 38, 82–87 (1978).

55. H. Reid, H. Marsden, Gonadal infiltration in children with leukaemia and lymphoma., J Clin
112
Pathol 33, 722 (1980).

56. I. Cunningham, The clinical behavior of 124 leukemic ovarian tumors: clues for improving
the poor prognosis, Leukemia Lymphoma 54, 1430–1436 (2012).

57. S. Zhang, W. Wang, Y. Kan, J. Liu, J. Yang, Extramedullary Infiltration of Acute
Lymphoblastic Leukemia in Multiple Organs on FDG PET/CT, Clin Nucl Med 43, 217–219
(2018).

58. Y. Lazebnik, What are the hallmarks of cancer?, Nat Rev Cancer 10, 232–233 (2010).

59. A. F. Chambers, A. C. Groom, I. C. MacDonald, Dissemination and growth of cancer cells in
metastatic sites, Nat Rev Cancer 2, 563–572 (2002).

60. P. J. Campbell, S. Yachida, L. J. Mudie, P. J. Stephens, E. D. Pleasance, L. A. Stebbings, L.
A. Morsberger, C. Latimer, S. McLaren, M.-L. Lin, D. J. McBride, I. Varela, S. A. Nik-Zainal,
C. Leroy, M. Jia, A. Menzies, A. P. Butler, J. W. Teague, C. A. Griffin, J. Burton, H. Swerdlow,
M. A. Quail, M. R. Stratton, C. Iacobuzio-Donahue, A. P. Futreal, The patterns and dynamics of
genomic instability in metastatic pancreatic cancer, Nature 467, 1109–1113 (2010).

61. M. Gerlinger, A. J. Rowan, S. Horswell, J. Larkin, D. Endesfelder, E. Gronroos, P. Martinez,
N. Matthews, A. Stewart, P. Tarpey, I. Varela, B. Phillimore, S. Begum, N. Q. nald, A. Butler,
D. Jones, K. Raine, C. Latimer, C. R. Santos, M. Nohadani, A. C. Eklund, B. Spencer-Dene, G.
Clark, L. Pickering, G. Stamp, M. Gore, Z. Szallasi, J. Downward, A. P. Futreal, C. Swanton,
Intratumor Heterogeneity and Branched Evolution Revealed by Multiregion Sequencing, New
Engl J Med 366, 883–892 (2012).

62. W. Liu, S. Laitinen, S. Khan, M. Vihinen, J. Kowalski, G. Yu, L. Chen, C. M. Ewing, M. A.
Eisenberger, M. A. Carducci, W. G. Nelson, nivasan Yegnasubramanian, J. Luo, Y. Wang, J. Xu,
W. B. Isaacs, T. Visakorpi, S. G. Bova, Copy number analysis indicates monoclonal origin of
lethal metastatic prostate cancer, Nat Med 15, 559–565 (2009).
113

63. N. Navin, J. Kendall, J. Troge, P. Andrews, L. Rodgers, J. McIndoo, K. Cook, A. Stepansky,
D. Levy, D. Esposito, L. Muthuswamy, A. Krasnitz, R. W. McCombie, J. Hicks, M. Wigler,
Tumour evolution inferred by single-cell sequencing, Nature 472, 90–94 (2011).

64. S. P. Shah, R. D. Morin, J. Khattra, L. Prentice, T. Pugh, A. Burleigh, A. Delaney, K.
Gelmon, R. Guliany, J. Senz, C. Steidl, R. A. Holt, S. Jones, M. Sun, G. Leung, R. Moore, T.
Severson, G. A. Taylor, A. E. Teschendorff, K. Tse, G. Turashvili, R. Varhol, R. L. Warren, P.
Watson, Y. Zhao, C. Caldas, D. Huntsman, M. Hirst, M. A. Marra, S. Aparicio, Mutational
evolution in a lobular breast tumour profiled at single nucleotide resolution, Nature 461, 809–
813 (2009).

65. S. L. Carter, K. Cibulskis, E. Helman, A. McKenna, H. Shen, T. Zack, P. W. Laird, R. C.
Onofrio, W. Winckler, B. A. Weir, R. Beroukhim, D. Pellman, D. A. Levine, E. S. Lander, M.
Meyerson, G. Getz, Absolute quantification of somatic DNA alterations in human cancer, Nat
Biotechnol 30, 413–421 (2012).

66. S. Nik-Zainal, L. B. Alexandrov, D. C. Wedge, P. Loo, C. D. Greenman, K. Raine, D. Jones,
J. Hinton, J. Marshall, L. A. Stebbings, A. Menzies, S. Martin, K. Leung, L. Chen, C. Leroy, M.
Ramakrishna, R. Rance, K. Lau, L. J. Mudie, I. Varela, D. J. McBride, G. R. Bignell, S. L.
Cooke, A. Shlien, J. Gamble, I. Whitmore, M. Maddison, P. S. Tarpey, H. R. Davies, E.
Papaemmanuil, P. J. Stephens, S. McLaren, A. P. Butler, J. W. Teague, G. Jönsson, J. E. Garber,
D. Silver, P. Miron, A. Fatima, S. Boyault, A. Langerød, A. Tutt, J. Martens, S. Aparicio, Å.
Borg, A. Salomon, G. Thomas, A.-L. Børresen-Dale, A. L. Richardson, M. S. Neuberger, A. P.
Futreal, P. J. Campbell, M. R. Stratton, B. the of the Consortium, Mutational Processes Molding
the Genomes of 21 Breast Cancers, Cell 149, 979–993 (2012).

67. G. M. Crane, E. Jeffery, S. J. Morrison, Adult haematopoietic stem cell niches, Nat Rev
Immunol 17, 573–590 (2017).

68. X. Sun, G. Cheng, M. Hao, J. Zheng, X. Zhou, J. Zhang, R. S. Taichman, K. J. Pienta, J.
114
Wang, CXCL12 / CXCR4 / CXCR7 chemokine axis and cancer progression, Cancer Metast Rev
29, 709–722 (2010).

69. M. V. Barbolina, M. Kim, Y. Liu, J. Shepard, A. Belmadani, R. J. Miller, L. D. Shea, S. M.
Stack, Microenvironmental Regulation of Chemokine (C-X-C-Motif) Receptor 4 in Ovarian
Carcinoma, Mol Cancer Res 8, 653–664 (2010).

70. A. Colmone, M. Amorim, A. L. Pontier, S. Wang, E. Jablonski, D. A. Sipkins, Leukemic
Cells Create Bone Marrow Niches That Disrupt the Behavior of Normal Hematopoietic
Progenitor Cells, Science 322, 1861–1865 (2008).

71. D. A. Sipkins, X. Wei, J. W. Wu, J. M. Runnels, D. Côté, T. K. Means, A. D. Luster, D. T.
Scadden, C. P. Lin, In vivo imaging of specialized bone marrow endothelial microdomains for
tumour engraftment., Nature 435, 969–73 (2005).

72. J. Juarez, K. Bradstock, D. Gottlieb, L. Bendall, Effects of inhibitors of the chemokine
receptor CXCR4 on acute lymphoblastic leukemia cells in vitro., Leukemia 17, 1294–300 (2003).

73. W. Shen, L. Bendall, D. Gottlieb, K. Bradstock, The chemokine receptor CXCR4 enhances
integrin-mediated in vitro adhesion and facilitates engraftment of leukemic precursor-B cells in
the bone marrow, (2001), doi:10.1016/s0301-472x(01)00741-x .

74. D. Gilbert, I. Chandler, A. McIntyre, N. Goddard, R. Gabe, R. Huddart, J. Shipley, Clinical
and biological significance of CXCL12 and CXCR4 expression in adult testes and germ cell
tumours of adults and adolescents, J Pathology 217, 94–102 (2009).

75. R. Kersseboom, S. Middendorp, G. M. Dingjan, K. Dahlenborg, M. Reth, H. Jumaa, R. W.
Hendriks, Bruton’s Tyrosine Kinase Cooperates with the B Cell Linker Protein SLP-65 as a
Tumor Suppressor in Pre-B Cells, J Exp Medicine 198, 91–98 (2003).

76. N. Feldhahn, P. Río, B. Soh, S. Liedtke, M. Sprangers, F. Klein, P. Wernet, H. Jumaa, W.-K.
115
Hofmann, H. Hanenberg, J. D. Rowley, M. Müschen, Deficiency of Bruton’s tyrosine kinase in
B cell precursor leukemia cells, P Natl Acad Sci Usa 102, 13266–13271 (2005).

77. D. Rawlings, D. Saffran, S. Tsukada, D. Largaespada, J. Grimaldi, L. Cohen, Mohr, J. Bazan,
M. Howard, N. Copeland, al. et, Mutation of unique region of Bruton’s tyrosine kinase in
immunodeficient XID mice, Science 261, 358–361 (1993).

78. M. Spaargaren, E. A. Beuling, M. L. Rurup, H. P. Meijer, M. D. Klok, S. Middendorp, R. W.
Hendriks, S. T. Pals, The B Cell Antigen Receptor Controls Integrin Activity through Btk and
PLCγ2, J Exp Medicine 198, 1539–1550 (2003).

79. M. F. de Rooij, A. Kuil, C. R. Geest, E. Eldering, B. Y. Chang, J. J. Buggy, S. T. Pals, M.
Spaargaren, The clinically active BTK inhibitor PCI-32765 targets B-cell receptor– and
chemokine-controlled adhesion and migration in chronic lymphocytic leukemia, Blood 119,
2590–2594 (2012).

80. SEER Cancer Stat Facts: Acute Lymphocytic Leukemia, (2019).

81. C.-H. H. Pui, L. L. Robison, A. Look, Acute lymphoblastic leukaemia., Lancet 371, 1030–43
(2008).

82. A. K. Fielding, A. Fielding, The Treatment of Adults with Acute Lymphoblastic Leukemia,
Ash Educ Program Book 2008, 381–389 (2008).

83. S. Farber, L. K. Diamond, R. D. Mercer, R. F. Sylvester, J. A. Wolff, Temporary Remissions
in Acute Leukemia in Children Produced by Folic Acid Antagonist, 4-Aminopteroyl-Glutamic
Acid (Aminopterin), New Engl J Medicine 238, 787–793 (1948).

84. E. Frei, E. J. Freireich, E. Gehan, D. Pinkel, J. F. Holland, O. Selawry, F. Haurani, C. L.
Spurr, D. M. Hayes, W. G. James, H. Rothberg, B. D. Sodee, W. R. Rundles, L. R. Schroeder, B.
Hoogstraten, I. J. Wolman, D. Traggis, T. Cooper, B. Gendel, F. Ebaugh, R. Taylor, Studies of
116
Sequential and Combination Antimetabolite Therapy in Acute Leukemia: 6-Mercaptopurine and
Methotrexate, Blood 18, 431–454 (1961).

85. L. Ding, T. J. Ley, D. E. Larson, C. A. Miller, D. C. Koboldt, J. S. Welch, J. K. Ritchey, M.
A. Young, T. Lamprecht, M. D. McLellan, J. F. McMichael, J. W. Wallis, C. Lu, D. Shen, C. C.
Harris, D. J. Dooling, R. S. Fulton, L. L. Fulton, K. Chen, H. Schmidt, J. Kalicki-Veizer, V. J.
Magrini, L. Cook, S. D. McGrath, T. L. Vickery, M. C. Wendl, S. Heath, M. A. Watson, D. C.
Link, M. H. Tomasson, W. D. Shannon, J. E. Payton, S. Kulkarni, P. Westervelt, M. J. Walter, T.
A. Graubert, E. R. Mardis, R. K. Wilson, J. F. DiPersio, Clonal evolution in relapsed acute
myeloid leukaemia revealed by whole-genome sequencing., Nature 481, 506–10 (2012).

86. M. Greaves, C. C. Maley, Clonal evolution in cancer, Nature 481, 306–313 (2012).

87. M. Shackleton, E. Quintana, E. R. Fearon, S. J. Morrison, Heterogeneity in cancer: cancer
stem cells versus clonal evolution., Cell 138, 822–9 (2009).

88. H. G. Einsiedel, A. von Stackelberg, R. Hartmann, R. Fengler, M. Schrappe, G. Janka-
Schaub, G. Mann, K. Hählen, U. Göbel, T. Klingebiel, W.-D. D. Ludwig, G. Henze, Long-term
outcome in children with relapsed ALL by risk-stratified salvage therapy: results of trial acute
lymphoblastic leukemia-relapse study of the Berlin-Frankfurt-Münster Group 87., Journal of
clinical oncology : official journal of the American Society of Clinical Oncology 23, 7942–50
(2005).

89. G. Buchanan, G. Rivera, B. Pollock, J. Boyett, A. Chauvenet, H. Wagner, D. Maybee, W.
Crist, D. Pinkel, Alternating drug pairs with or without periodic reinduction in children with
acute lymphoblastic leukemia in second bone marrow remission: a Pediatric Oncology Group
Study., Cancer 88, 1166–74 (2000).

90. S. Lawson, G. Harrison, S. Richards, A. Oakhill, R. Stevens, O. Eden, P. Darbyshire, The
UK experience in treating relapsed childhood acute lymphoblastic leukaemia: a report on the
medical research council UKALLR1 study., British journal of haematology 108, 531–43 (2000).
117

91. J. A. Meyer, J. Wang, L. E. Hogan, J. J. Yang, S. Dandekar, J. P. Patel, Z. Tang, P. Zumbo,
S. Li, J. Zavadil, R. L. Levine, T. Cardozo, S. P. Hunger, E. A. Raetz, W. E. Evans, D. J.
Morrison, C. E. Mason, W. L. Carroll, Relapse-specific mutations in NT5C2 in childhood acute
lymphoblastic leukemia, Nature Genetics 45, 290–294 (2013).

92. G. Tzoneva, A. Perez-Garcia, Z. Carpenter, H. Khiabanian, V. Tosello, M. Allegretta, E.
Paietta, J. Racevskis, J. M. Rowe, M. S. Tallman, M. Paganin, G. Basso, J. Hof, R. Kirschner-
Schwabe, T. Palomero, R. Rabadan, A. Ferrando, Activating mutations in the NT5C2
nucleotidase gene drive chemotherapy resistance in relapsed ALL., Nature medicine 19, 368–71
(2013).

93. X. Ma, M. Edmonson, D. Yergeau, D. M. Muzny, O. A. Hampton, M. Rusch, G. Song, J.
Easton, R. C. Harvey, D. A. Wheeler, J. Ma, H. Doddapaneni, B. Vadodaria, G. Wu, P.
Nagahawatte, W. L. Carroll, I.-M. M. Chen, J. M. Gastier-Foster, M. V. Relling, M. A. Smith,
M. Devidas, J. M. Auvil, J. R. Downing, M. L. Loh, C. L. Willman, D. S. Gerhard, C. G.
Mullighan, S. P. Hunger, J. Zhang, Rise and fall of subclones from diagnosis to relapse in
pediatric B-acute lymphoblastic leukaemia., Nature communications 6, 6604 (2015).

94. B. Szymanska, U. Wilczynska-Kalak, M. H. Kang, N. L. Liem, H. Carol, I. Boehm, D.
Groepper, P. C. Reynolds, C. F. Stewart, R. B. Lock, Pharmacokinetic Modeling of an Induction
Regimen for In Vivo Combined Testing of Novel Drugs against Pediatric Acute Lymphoblastic
Leukemia Xenografts, PLoS ONE 7, e33894 (2012).

95. E. Park, E. Gang, Y.-T. Hsieh, P. Schaefer, S. Chae, L. Klemm, S. Huantes, M. Loh, E. M.
Conway, E.-S. Kang, H. Koo, W.-K. Hofmann, N. Heisterkamp, L. Pelus, G. Keerthivasan, J.
Crispino, M. Kahn, M. Müschen, Y.-M. Kim, Targeting survivin overcomes drug resistance in
acute lymphoblastic leukemia, Blood 118, 2191–9 (2011).

96. E. Tijchon, L. van Emst, L. Yuniati, D. van Schenau, J. Havinga, J.-P. Rouault, P. M.
Hoogerbrugge, F. N. van Leeuwen, B. Scheijen, Tumor suppressors BTG1 and BTG2 regulate
118
early mouse B-cell development, Haematologica 101, e272–e276 (2016).

97. M. Islaih, B. W. Halstead, I. A. Kadura, B. Li, J. L. Reid-Hubbard, L. Flick, J. L. Altizer, T.
J. Deahl, D. K. Monteith, R. K. Newton, D. E. Watson, Relationships between genomic, cell
cycle, and mutagenic responses of TK6 cells exposed to DNA damaging chemicals, Mutat Res
Fundam Mol Mech Mutagen 578, 100–116 (2005).

98. J.-G. Chen, C.-P. Yang, M. Cammer, S. Horwitz, Gene expression and mitotic exit induced
by microtubule-stabilizing drugs., Cancer Res 63, 7891–9 (2003).

99. N. W. van de Donk, M. L. Janmaat, T. Mutis, J. J. van Bueren, T. Ahmadi, K. A. Sasser, H.
M. Lokhorst, P. W. Parren, Monoclonal antibodies targeting CD38 in hematological
malignancies and beyond, Immunol Rev 270, 95–112 (2016).

100. J. Delgado, J. G. Bustos, M. Jimenez, E. Quevedo, F. Hernandez-Navarro, Are Activation
Markers (CD25, CD38 and CD103) Predictive of Sensitivity to Purine Analogues in Patients
with T-cell Prolymphocytic Leukemia and other Lymphoproliferative Disorders?, Leukemia
Lymphoma 43, 2331–2334 (2009).

101. Y. Yano, E. Shiba, J. Kambayashi, kon, T. Kawasaki, K. Fujitani, J. Kang, T. Mori, The
effects of calpeptin (a calpain specific inhibitor) on agonist induced microparticle formation from
the platelet plasma membrane., Thrombosis research 71, 385–96 (1993).

102. A. P. Atanassoff, H. Wolfmeier, R. Schoenauer, A. Hostettler, A. Ring, A. Draeger, E. B.
Babiychuk, Microvesicle shedding and lysosomal repair fulfill divergent cellular needs during
the repair of streptolysin O-induced plasmalemmal damage., PloS one 9, e89743 (2014).

103. S. Jorfi, E. A. Ansa-Addo, S. Kholia, D. Stratton, S. Valley, S. Lange, J. Inal, Inhibition of
microvesiculation sensitizes prostate cancer cells to chemotherapy and reduces docetaxel dose
required to limit tumor growth in vivo., Scientific reports 5, 13006 (2015).

119
104. A. D. Boiko, S. Porteous, O. V. Razorenova, V. I. Krivokrysenko, B. R. Williams, A. V.
Gudkov, A systematic search for downstream mediators of tumor suppressor function of p53
reveals a major role of BTG2 in suppression of Ras-induced transformation, Gene Dev 20, 236–
252 (2006).

Abstract (if available)

Abstract Cellular heterogeneity is a cause of treatment resistance in cancer

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Dissecting the heterogeneity of mouse hematopoietic stem cells in vivo

PDF

Investigating molecular roadblocks to enhance direct cellular reprogramming

PDF

Tracking human acute lymphoblastic leukemia cell clones in xenograft mouse models

PDF

Functional compensation between hematopoietic stem cell clones in vivo

PDF

Deciphering heterogeneity of preleukemic clonal expansion

PDF

The role of ERK1/2 in mouse embryonic stem cell fate control

PDF

Generation of patterned and functional kidney organoids that recapitulate the adult kidney collecting duct system

PDF

Fibroblastic connective tissue cells: the blastema stem cells and source of large-scale chondrogenesis in the regenerating lizard tail

PDF

The effect of cytomegalovirus on gene expression of pediatric acute lymphoblastic leukemia

PDF

Identification of therapeutic targets for neurons and microglia in amyotrophic lateral sclerosis

PDF

Molecular programs in epithelial morphogenesis in mammalian organ systems

PDF

Role of beta-catenin in mouse epiblast stem cell, embryonic stem cell self-renewal and differentiation

PDF

Derivation of cardiomyocyte-propelled motile cellular constructs via de novo developmental trajectory

PDF

Neural crest-derived cranial pericytes and their dysfunction in disease

PDF

Clinical outcomes of allogeneic hematopoietic stem cell transplant in acute lymphoblastic leukemia patients: a quality improvement project and systematic review meta-analysis

PDF

Role of STAT3 phosphorylation in mouse embryonic stem cell self-renewal and differentiation

PDF

Transcriptional and epigenetic mechanisms underlying sensory hair cell differentiation and regeneration

PDF

Generation and characterization of an in vitro model of inner ear sensory hair cells using transcription factor mediated cellular reprogramming

PDF

Discovery of a novel role for Sonic Hedgehog during the early stages of large-scale murine rib regeneration

PDF

Molecular basis of mouse epiblast stem cell and human embryonic stem cell self‐renewal

Asset Metadata

Creator Contreras-Trujillo, Humberto (author)

Core Title Molecular signatures underlying intercellular differences in leukemia progression and chemotherapy response

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Development, Stem Cells and Regenerative Medicine

Publication Date 02/06/2020

Defense Date 12/16/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag acute lymphoblastic leukemia,Blood,cancer,chemotherapy,clonal evolution,clonal tracking,heterogeneity,leukemia,mRNA sequencing,OAI-PMH Harvest,oncology,patient derived xenograft,single cell

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ying, Qi-long (committee chair), Lu, Rong (committee member), Yu, Min (committee member)

Creator Email hcontrer@usc.edu,HCTrujillo@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-266283

Unique identifier UC11673254

Identifier etd-ContrerasT-8154.pdf (filename),usctheses-c89-266283 (legacy record id)

Legacy Identifier etd-ContrerasT-8154.pdf

Dmrecord 266283

Document Type Dissertation

Rights Contreras-Trujillo, Humberto

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA