Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Development of a colorectal cancer-on-chip to investigate the tumor microenvironment's role in cancer progression
(USC Thesis Other)
Development of a colorectal cancer-on-chip to investigate the tumor microenvironment's role in cancer progression
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Development of a colorectal cancer-on-chip to investigate the tumor microenvironment’s role in
cancer progression
by
Carly Strelez
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Cancer Biology and Genomics)
May 2022
ii
ACKNOWLEDGMENTS
I am extremely grateful for the opportunity to complete this research project at the
Lawrence J. Ellison Institute for Transformative Medicine (EITM) at USC. The level of support
for such a risky scientific project, of intellectual stimulation, and of comradery from the entire
Ellison Institute was unmatched. This is particularly true of my mentor, Dr. Shannon
Mumenthaler. Thank you for seeing something in me that I wasn’t quite sure existed, for giving
me the opportunity to rise to the occasion, for never once doubting me, and for being my biggest
cheerleader. It has shaped both the scientist and the person I am today and there will never be
enough words or space to describe the magnitude of the role you played in my life.
This project would not have been possible without the generous support of the many
EITM doners. In addition, a majority of this work was funded by the NCI Tissue Engineering
Consortium RO1 CA241138-01A, a Stop Cancer Grant, and the USC Norris Comprehensive
Cancer Center Core Grant P30CA014089 awarded to Shannon M. Mumenthaler.
I am fortunate to have the support of my incredible friends and family. There are more
people within my circle than I could possibly name, but to all my friends across all my life stages
(Connecticut, Gettysburg College, Washington, D.C., USC, and Los Angeles): thank you for
always listening, for teaching me something about myself and the world, for ignoring
responsibilities when I request a last-minute happy hour, for having a place for me to sleep
when I visit and need to travel on a budget, and for always helping me get back on my feet
when life gets tough. To my immediate and extended family (Mom, Dad, Ryan, and Annie):
thank you for never once questioning my desire to pick up and move across the country to an
unknown city to pursue such an unknown degree. Anything I have ever accomplished is
because of your love, choices, and sacrifices.
iii
TABLE OF CONTENTS
Acknowledgements ....................................................................................................................... ii
List of Tables ................................................................................................................................ v
List of Figures ............................................................................................................................... vi
Abstract ....................................................................................................................................... vii
Introduction .................................................................................................................................... 1
Chapter 1 ....................................................................................................................................... 3
1.1 Introduction .................................................................................................................. 3
1.2 Results ......................................................................................................................... 4
CRC-Chip Overview .................................................................................................................................. 4
Development and Characterization of CRC-Chip ...................................................................................... 5
Metabolic Comparison of Intestine Chips versus CRC-Chips ................................................................. 11
Examination of Tumor Cell Intravasation ................................................................................................. 25
CRC Cells Exhibit Phenotypic Heterogeneity During Intravasation ......................................................... 29
Initial Characterization of the Role of Mechanical and Biochemical Cues from the TME on CRC Invasion
................................................................................................................................................................. 32
Optimization of Liver Chips to Study CRC Extravasation ........................................................................ 35
1.3 Discussion ................................................................................................................. 40
1.4 Future Directions ....................................................................................................... 44
1.5 Methods ..................................................................................................................... 44
Chapter 2 ..................................................................................................................................... 58
2.1 Introduction ................................................................................................................ 58
2.2 Results ....................................................................................................................... 60
CRC CAF Characterization ..................................................................................................................... 60
CAFs Promote or Restrain Invasion in a Patient-Specific Manner .......................................................... 63
Gene Expression Analysis Suggest CAFs Show Patient- and Organ-Specific Heterogeneity ................ 66
Colon CAFs Show Changes in Neurotransmitter Signaling Pathways .................................................... 69
2.3 Discussion ................................................................................................................. 70
iv
2.4 Future Directions ....................................................................................................... 73
2.5 Methods ..................................................................................................................... 74
Chapter 3 ..................................................................................................................................... 77
3.1 Introduction ................................................................................................................ 77
3.2 Results ....................................................................................................................... 78
CRC Tumor Cells Invasion Increases in the Presence of Mechanical, Peristalsis-like Forces ............... 78
CRC Tumor Cells Secrete GABA in Response to Peristalsis .................................................................. 82
Peristalsis-Mediated CRC Tumor Cell Invasion is GABA R A-Mediated ................................................... 85
Invaded, Circulating CRC Tumor Cells from Stretched CRC-Chips Show Increased Adhesive Properties
and can Colonize the Liver ...................................................................................................................... 88
3.3 Discussion ................................................................................................................. 90
3.4 Future Directions ....................................................................................................... 92
3.5 Methods ..................................................................................................................... 93
Chapter 4 ..................................................................................................................................... 98
Bibliography ............................................................................................................................... 100
v
LIST OF TABLES
Introduction
In vitro model systems ................................................................................................................... 2
Chapter 1
Table 1.1. Chip locations specified for cell types and ECM composition used in the CRC-Chip . 9
Table 1.2. Differential metabolites between D0 and D6 of epithelial channel effluent for the
HCT116-CRC-Chip. Related to Figure 1.6 .................................................................................. 14
Table 1.3. Differential metabolites between D0 and D6 of endothelial channel effluent for the
HCT116-CRC-Chip, Related to Figure 1.6 .................................................................................. 17
Table 1.4 Differential metabolites between D0 and D6 of epithelial channel effluent for the
HT29-CRC-Chip, Related to Figure 1.6 ...................................................................................... 18
Table 1.5. Differential metabolites between D0 and D6 of endothelial channel effluent for the
HT29-CRC-Chip, Related to Figure 1.6 ...................................................................................... 21
Table 1.6. Differential metabolites between the Intestine Chip and the HCT116-CRC-Chip
epithelial effluent on D6, Related to Figure 1.6 ........................................................................... 21
Table 1.7. Top significantly altered metabolic pathways from IPA analysis of differential
metabolites between Intestine Chip and HCT116-CRC-Chip epithelial effluent on D6. Related to
Figure 1.6 .................................................................................................................................... 24
Table 1.8 Top significantly altered metabolic pathways from IPA analysis of differential
metabolites between Intestine Chip and HCT116-CRC-Chip endothelial effluent on D6. Related
to Figure 1.7 ................................................................................................................................ 27
Table 1.9. Chip locations specified for cell types and ECM composition used in the Liver-Chip 35
Table 1.10. Gene specific primers for qPCR. Related to Figure 1.10 ......................................... 46
Chapter 2
Table 2.1. Clinical details of patient samples .............................................................................. 61
Table 2.2. Enriched pathways ..................................................................................................... 68
Table 2.3. Gene specific primers for qPCR, related to Figure 2.1 ............................................... 75
Chapter 3
Table 3.1. Fold changes of neurotransmitter intensities in the epithelial channel effluent in
response to peristalsis as measured by mass spectrometry-based metabolomics .................... 83
Table 3.2. Gene specific primers for qPCR. Related to Figures 3.1 ........................................... 95
vi
LIST OF FIGURES
Chapter 1
Figure 1.1. CRC-Chip can be multiplexed to interrogate tumor cell behavior ............................... 5
Figure 1.2. Experimental timeline .................................................................................................. 5
Figure 1.3. Caco2 Intestine Chip exhibits different cell types present in the human intestine ...... 6
Figure 1.4. CRC-Chip tissue structure .......................................................................................... 8
Figure 1.5. Tight junction formation and tumor cell morphology in the HCT116-CRC-Chip ........ 10
Figure 1.6. Metabolic analyses validate use of CRC-Chip to model CRC progression ............... 13
Figure 1.7. Validation of CRC tumor cell invasion from an epithelial to endothelial compartment,
mimicking intravasation ............................................................................................................... 26
Figure 1.8. Validation of CRC-Chip intravasation assay ............................................................. 29
Figure 1.9. HT29 and HCT116 cells show different expression of epithelial and mesenchymal
markers ....................................................................................................................................... 30
Figure 1.10. CRC cell heterogeneity during intravasation on-chip .............................................. 31
Figure 1.11. The TME influences tumor cell invasion ................................................................. 33
Figure 1.12. Characterization of human Liver Chip ..................................................................... 36
Figure 1.13. Schematic of CRC tumor cell extravasation on-chip assay .................................... 38
Figure 1.14. CRC Tumor Cells Extravasate on a Healthy Human Liver Chip ............................. 39
Chapter 2
Figure 2.1 CAF Plasticity Schematic ........................................................................................... 60
Figure 2.2. Patient-derived CRC CAFs show heterogeneous expression of CAF markers ........ 62
Figure 2.3. CAF CM shows heterogeneous cytokine profiles and influences tumor cell invasion
in a patient-specific manner ........................................................................................................ 65
Figure 2.4. Physical interactions between CAFs and tumor cells promote or restrain tumor cell
invasion in a patient-specific manner .......................................................................................... 66
Figure 2.5. CAFs show heterogeneous gene expression across the primary tumor and the liver
metastatic site ............................................................................................................................. 67
Figure 2.6. Upregulated genes in CAFs ...................................................................................... 69
Figure 2.7. Pathway analysis indicate neurotransmitter signaling is changed within colon CAFs
.................................................................................................................................................... 70
Chapter 3
Figure 3.1. CRC tumor cells increase invasion and decrease proliferation in response to
peristalsis .................................................................................................................................... 80
Figure 3.2. CRC tumor cells respond dynamically to mechanical force perturbations ................ 81
Figure 3.3. On-chip characterization of barrier function and proliferation in response to
peristalsis .................................................................................................................................... 82
Figure 3.4. CRC tumor cells secrete GABA in response to mechanical, peristalsis-like stretching
.................................................................................................................................................... 85
Figure 3.5. CRC tumor cells respond to peristalsis through GABA RA signaling ......................... 87
Figure 3.6. Peristalsis induces adhesion changes in CRC CTCs, resulting in increased
extravasation in a healthy human liver model ............................................................................. 89
vii
ABSTRACT
Colorectal cancer (CRC) is one of the deadliest cancers in the U.S., yet we still
understand very little about the mechanisms behind this disease. Therefore, I am developing a
CRC-Chip model that recapitulates the complex nature of progression to increase our
understanding of CRC and accelerate the discovery of new treatments. The organ-on-chip
technology developed by Emulate, Inc. maintains physiologically relevant aspects of organ
structure and function by incorporating tissue compartments and mechanical forces to mimic in
vivo peristalsis and fluid flow. The CRC-Chip consists of two microfluidic compartments
separated by a porous membrane, with endothelial cells in the bottom channel and normal colon
epithelial cells plus fluorescently-labeled CRC cell lines in the top channel. I optimized various
iterations of the CRC-Chip to include CRC cell lines, the addition of patient-derived cancer
associated fibroblasts (CAFs), and patient-derived CRC organoids. Several assays were
developed to monitor cancer progression: an imaging-based invasion assay in which cancer
cells are visualized traversing from the epithelial compartment into the blood vessel
compartment via confocal microscopy; and a mass spectrometry-based metabolomics analysis
of the effluent to monitor changes in metabolites over time. In this dissertation I show that 1) the
CRC-Chip recapitulates the varying aggressiveness of colon cancers and cancer biology, 2) the
CRC-Chip allows for studying cancer associated fibroblast (CAF) functional heterogeneity and
3) peristalsis-like mechanical forces influences the invasive capabilities of tumor cells through
GABAergic signaling, something that could not previously be studied. The work described here
with a unique model system sets the stage for identifying new therapies targeting the stroma
and cancer cells responding to a mechanical microenvironment.
1
INTRODUCTION
Cancer is a highly complex adaptive system, which makes it challenging to study.
Despite significant advances in drug development and translational research, we still
understand very little about the mechanisms behind this disease. Colorectal cancer (CRC) is
one of the deadliest cancers worldwide with over 900,000 people dying from the disease each
year (Siegel et al., 2021). In the United States, the 5-year survival rate for patients with
metastatic CRC is less than 15% (Siegel et al., 2020). To address this dismal outcome, there is
an urgent need to better understand and ultimately control aspects of cancer progression.
Tumor metastasis is an elaborate cascade of events whereby cells from the primary tumor
invade the surrounding tissue and intravasate into the blood or lymphatic vessels, extravasate
into a distant organ, evade the immune system, and grow into secondary, metastatic tumors
(Batista et al., 2019).
Genetically engineered mouse models and patient-derived xenografts have been critical
in advancing the tumor metastasis field by providing an in vivo system that can model tumor
spread from primary to distant organ sites through the vascular network (Bürtin et al., 2020;
Walrath et al., 2010). However, limitations remain with traditional preclinical cancer models,
including a lack of in vitro model systems that imitate human physiology, specifically aspects of
the tumor microenvironment (TME), and the inability of in vivo animal experiments to
recapitulate and tune human organ microenvironments (Gould et al., 2015; Ledford, 2011; Mak
et al., 2014). Recent advances in in vitro culture systems have overcome some of these
limitations. In particular, 3D microfluidic organ-on-chip (OOC) systems incorporate unique
features to better model in vivo cell-cell and cell-extracellular matrix (ECM) interactions (Bhatia
and Ingber, 2014; Sarvestani et al., 2020), which can support novel interrogations of cancer
progression (outlined in table format below).
2
In vitro model systems.
System Type Disadvantages Advantages
2D Systems
Single Cell Types Simple; Doesn’t
translate to human
biology
High-throughput;
applicable to drug
screening
Heterocellular
cultures
Missing spatial
information; difficult to
study time dynamics
Amenable to high-
throughput assays
Transwell Oversimplifying a
complex process
Simple
3D Systems
Organoids Expensive; variable;
assays are less
developed
Recapitulates patient
tumor; applicable to
drug screening
Organ-on-chip Expensive; assays are
less developed;
technically challenging
Tunable; mechanical
forces can be studied;
potential to study a
wide range of cancer
mechanisms
This dissertation focuses on the development of a colorectal cancer-on-chip (CRC-Chip)
to study early metastatic events and to interrogate how aspects of the TME, specifically CAFs
and mechanical forces, influence CRC progression. While the field is slowly beginning to
understand how the TME impacts cancer metastasis, there is an urgent need to develop
tunable, physiologically-relevant in vitro model systems that can provide insight on the functional
implications of the TME on cancer progression. This work contributes to the cancer field by 1)
demonstrating how Organ Chips can be used to provide novel mechanistic knowledge of cancer
biology, 2) expanding the known role of CAF heterogeneity in CRC progression, and 3)
introducing the hypothesis that mechanical forces increase the invasive capabilities of CRC
tumor cells.
3
CHAPTER 1
Development of Organ-on-Chip models to study CRC progression
Associated publications:
Emma J. Fong, Carly Strelez, and Shannon M. Mumenthaler. A perspective on expanding our
understanding of cancer treatments by integrating approaches from the biological and physical
sciences. SLAS Discovery 2020.
Carly Strelez, Sujatha Chilakala, Kimya Ghaffarian, Roy Lau, Erin Spiller, Nolan Ung, Danielle
Hixon, Ah Young Yoon, Ren X. Sun, Heinz-Josef Lenz, Jonathan E. Katz, and Shannon M.
Mumenthaler. Human colorectal cancer-on-chip model to study the microenvironmental influence
on early metastatic spread. iScience 2021.
Carly Strelez, Kimya Ghaffarian, and Shannon M. Mumenthaler. Multiplexed imaging and effluent
analysis to monitor cancer cell intravasation using a colorectal cancer-on-chip. In press at STAR
Protocols 2021.
Contributions:
All CRC-Chip and Liver-Chip experiments were optimized by C. Strelez based on expertise and
advice from Emulate, Inc. CRC-Chip and Liver-Chip experiments were completed by C. Strelez.
C. Strelez also completed qPCRs, effluent analyses, and invasion assays. Mass spectrometry-
based metabolomic analyses were performed by S. Chilakala, A.Y. Yoon, and J.E. Katz.
1.1 INTRODUCTION
OOCs are designed to model normal or diseased organ-level structure and function by
incorporating tissue compartments and physical forces that mimic in vivo cyclic strain (i.e.,
peristalsis-like motions) and fluid shear stress (Basson, 2007; Gayer and Basson, 2009). The
microfluidic nature of these systems sustains longer-term experiments and allows for continuous
effluent collection to monitor byproducts as an indirect measure of tissue function and viability
(Bai et al., 2015; Jang et al., 2019; McAleer et al., 2019; Pavesi et al., 2017). When combined
with imaging-based approaches (Lee et al., 2018; Pavesi et al., 2017; Ying et al., 2015), OOCs
support dynamic cell phenotyping in a non-invasive manner. OOC models across a variety of
cancer types (e.g., breast, lung, colon, and pancreatic cancers) have been engineered to
4
interrogate important biological processes in cancer, such as angiogenesis, epithelial-
mesenchymal transition, cancer cell metastasis, and therapeutic response (Caballero et al.,
2017; Sontheimer-Phelps et al., 2019).
1.2 RESULTS
CRC-Chip Overview
My goal was to develop a CRC-Chip, integrating multiple in vivo-relevant cell types and
physical forces, to reveal how diverse tumor microenvironment (TME) cues work in concert to
influence the spread of a primary colon tumor. I introduced aspects of the TME in a step-wise
fashion to support the study of tumor-TME interactions in a tunable, physiologically relevant
system. Specifically, I expanded upon previous OOC models by incorporating physical forces to
mimic peristalsis, an important factor in colon physiology (Gayer and Basson, 2009) and host-
microbe interactions in the healthy intestine (Grassart et al., 2019; Kim et al., 2016). In addition,
I integrated aspects of the stromal TME, such as cancer-associated fibroblasts, which have
been implicated in metastatic spread (Quail and Joyce, 2013). This resulted in a heterocellular
tumor compartment interfaced with a blood vessel compartment to study CRC as a diseased
organ. I demonstrate that this model is suitable to investigate early stages of the CRC
metastatic process, mimicking the intravasation of tumor cells into a blood vessel, which can be
monitored via on-chip imaging and mass spectrometry-based metabolomics. As depicted in the
graphical abstract (Figures 1.1 and 1.2), the CRC-Chip model can be used to interrogate tumor
cell behavior in a “multiplexed” fashion. From a single chip, tumor cell morphology, growth rate,
and invasion dynamics can be monitored via confocal microscopy and frequent analyses of the
effluent, such as metabolomics or cell shedding, can be performed.
5
Figure 1.1. CRC-Chip can be multiplexed to interrogate tumor cell behavior.
Figure 1.2: Experimental timeline. Detailed overview of key steps on critical days of chip preparation, on-chip imaging,
and effluent collection.
Development and Characterization of CRC-Chip
An Intestine Chip, consisting of intestinal epithelial cells (Caco2 C2BBe1) and
endothelial cells (human umbilical vein endothelial cells; HUVECs) seeded in an ECM-coated
6
chip, has been previously developed to model the intestine (Jalili-Firoozinezhad et al., 2018;
Kim et al., 2012; Kim et al., 2016). On chip, the Caco2 C2BBe1 clone more closely resembles
the normal human colon due to the formation of a polarized monolayer of epithelial cells
displaying an apical brush border (Peterson and Mooseker, 1992). With fluid flow and cyclic
strain, the Caco2 C2BBe1 cells form 3D-like architecture and differentiate into the four main
intestinal cell lineages (goblet, enteroendocrine, Paneth, and enterocytes) (Figure 1.3). The
Intestine Chip emulates tissue function by displaying an intact intestinal barrier with the
endothelial cells forming vessel-like structures along the bottom channel (Jalili-Firoozinezhad et
al., 2018; Kim et al., 2012; Kim et al., 2016). I modified the Caco2 Intestine Chip to model CRC
by introducing epithelial and stromal cells (tumor cell lines or patient-derived tumor organoids
and patient-derived cancer associated fibroblasts (CAFs)) into the top channel (Figure 1.4A)
and endothelial cells (HUVEC) into the bottom channel. ECM composition, cell types, on-chip
locations, and fluorescent labels used to distinguish cell types are outlined in Table 1.1. Fluid
flow and cyclic, peristalsis-like mechanical deformations were introduced to complete the
physiologically-relevant epithelial:endothelial tissue:tissue interface and create a CRC-Chip
system.
Figure 1.3. Caco2 Intestine Chip exhibits different cell types present in the human intestine. Representative
confocal immunofluorescent images of Caco2 C2BBe1 cells on day 6 in the top epithelial channel of the Intestine
Chip stained for markers of Paneth cells (Lysozyme), absorptive cells (Sucrose Isomerase; SI), entero-endocrine
cells (Chromogranin A; CHGA) and mucus-secreting Goblet cells (Mucin 2; MUC2). Cell nuclei are labeled with DAPI
(blue). Scale bars represent 200 µm.
7
To generate the CRC-Chip, I first allowed the epithelium to form a monolayer, develop
villi-like structures, and establish a complete, functional barrier (approximately 2-3 days (Kim et
al., 2012; Kim et al., 2016)) before seeding CRC tumor cells. CRC tumor cells were seeded at a
low density relative to the Caco2 C2BBe1 epithelial cells (1:5 tumor:epithelial cell seeding) in
order to mimic cancer development within the colon epithelium. After the addition of tumor cells,
the chips were placed under constant flow (30 μL hr
-1
) and stretch conditions (10%
deformation; 0.2 Hz) for up to two weeks. The epithelial:endothelial tissue layers were visualized
by immunofluorescence staining of the endothelial cells (VE-cadherin) in the bottom
compartment and the Caco2 C2BBe1 (E-cadherin) and cancerous (H2B-GFP) epithelial cells in
the top compartment (Figure 1.4B). Clusters of HCT116 tumor cells were observed on top of the
3D structures formed by the Caco2 C2BBe1 cells. I found the addition of CRC tumor cell lines
did not noticeably impact the formation of tight junctions, as shown by immunofluorescence
staining of ZO-1 in the epithelial and endothelial channel (Figure1.4C). Large-scale images
show strong ZO-1 expression across the length of the epithelial channel in both the Caco2
Intestine Chips and the CRC-Chips seeded with HCT116 tumor cells (HCT116-CRC-Chip)
(Figure 1.5A&B). In addition, the presence of the HCT116 tumor cells did not significantly
change the ability of the Caco2 C2BBe1 cells to form a stable intestinal barrier over the course
of the experiments (Figure 1.4D). I modeled the development of CRC “hot spots” along the
length of the colon. On chips, the tumor cells grew in 3D clusters extending into the lumen
rather than the monolayer morphology seen on plastic (Figure 1.5C&D), suggesting the chip
structure can be used to mimic the progression of colon cancer in which a polyp forms in the
colonic crypts before eventually evolving into a cancerous lesion that grows into the intestinal
lumen (Dekker et al., 2019; Humphries and Wright, 2008).
8
Figure 1.4. CRC-Chip tissue structure. (A) The organ-on-chip platform (schematic courtesy of Emulate, Inc.)
consists of an epithelial channel (1) comprised of epithelial and cancerous cells (3) and an endothelial channel (2)
comprised of HUVEC cells (4) separated by a porous membrane (5). To model cell-cell interactions in the TME, the
CRC-Chip was modified to include layers of different cell types in the epithelial channel. CRC tumor cells were
seeded on top of the epithelial cells. A stromal layer, comprised of CAFs, can be incorporated into the epithelial
channel. (B) Confocal fluorescence images of a chip cross-section spanning 106 μm from the top of the endothelial
channel into the epithelial channel, highlighting the endothelial:epithelial tissue:tissue interface. HUVEC cells are
labeled with anti-VE cadherin (red). Caco2 C2BBe1 cells labeled with anti-E-Cadherin (purple) form 3D-like structures
in the top epithelial channel. HCT116 H2B-GFP cells grow in clusters on top of the Caco2 cells. Nuclei are labeled
with DAPI (blue). Scale bar is 100 μm. (C) Representative confocal immunofluorescent images of the epithelial (top)
and endothelial (bottom) channels of an Intestine Chip (left) and CRC-Chip (right) stained for ZO-1 (gold) on day 6.
DAPI (blue) labels the nuclei of the Caco2 C2BBe1 cells in the epithelial channel and HUVECs in the endothelial
channel. White arrows designate HCT116 (green) in the epithelial channel of the CRC-Chip. Scale bars represent
200 μm. Images are maximum projections that span a 15 μm Z-height in the epithelial channel and a 10 μm Z-height
in the endothelial channel with a 5 μm step size. (D) The apparent permeability (P app) of the intestinal epithelial cells
in the top channel was not changed when HCT116 tumor cells were added to the CRC-Chips. The concentration of
inulin-FITC that diffused from the epithelial channel to the endothelial channel was used to calculate P app (N=3
Chips). Data are represented as mean ± SEM and analyzed using a 2-way ANOVA; p>0.05.
chip (Figure 1D), with well-defined epithelial tight junctions, as demonstrated by ZO-1 protein stain-
ing and endothelial adherent junctions visualized using antibodies against VE-cadherin
(Dawson et al., 2016). Importantly, these culture conditions resulted in a time-dependent improve-
ment of intestinal permeability as indicated by the low permeability coefficient (Papp) of fluores-
cently labeled dextran recorded in the Duodenum Intestine-Chip generated from organoid-derived
cells of three different individuals (Figure 1E). Overall, this data indicates that the human adult Duo-
denum Intestine-Chip supports the formation of a functional barrier with in vivo relevant cytoarchi-
tecture, cell-cell interactions, and permeability parameters.
To confirm differentiation of the organoid-derived cells within the chip into all of the distrinct epi-
thelial cell lineages as found in vivo, we assessed average mRNA gene expression levels of cell-type-
Figure 1. Duodenum Intestine-Chip: a microengineered model of the human duodenum. (a) Brightfield images of
human duodenal organoids (top) and human microvascular endothelial cells (bottom) acquired before their
seeding into epithelial and endothelial channels of the chip, respectively. (b) Schematic representation of
Duodenum Intestine-Chip, including its top view (left) and vertical section (right) showing: the epithelial (1; blue)
and vascular (2; pink) cell culture microchannels populated by intestinal epithelial cells (3) and endothelial cells (4),
respectively, and separated by a flexible, porous, ECM-coated PDMS membrane (5). (c) Scanning electron
micrograph showing complex intestinal epithelial tissue architecture achieved by duodenal epithelium grown for 8
days on the chip (top) in the presence of constant flow of media (30 ml/hr) and cyclic membrane deformations (10%
strain, 0.2 Hz). High magnification of the apical epithelial cell surface with densely packed intestinal microvilli
(bottom). See Figure 1-figure supplement demonstrating the effect of mechanical forces on the cytoarchitecture of
epithelial cells and the formation of intestinal microvilli (d) Composite tile scan fluorescence image 8 days post-
seeding (top) showing a fully confluent monolayer of organoid-derived intestinal epithelial cells (magenta, ZO-1
staining) lining the lumen of Duodenum Intestine-Chip and interfacing with microvascular endothelium (green, VE-
cadherin staining) seeded in the adjacent vascular channel. Higher magnification views of epithelial tight junctions
(bottom left) stained against ZO-1 (magenta) and endothelial adherence junctions visualized by VE-cadherin
(bottom right) staining. Cells nuclei are shown in gray. Scale bars, 1000 mm (top), 100 mm (bottom) (e) Apparent
permeability values of Duodenum Intestine-Chips cultured in the presence of flow and stretch (30 ml/hr; 10% strain,
0.2 Hz) for up to 10 days. Papp values were calculated from the diffusion of 3 kDa Dextran from the luminal to the
vascular channel. Data represent three independent experiments performed with three different chips/donor, total
of three donors; Error bars indicate s.e.m.
The online version of this article includes the following figure supplement(s) for figure 1:
Figure supplement 1. Flow-induced increase in primary intestinal epithelial cells height and microvilli formation.
Kasendra et al. eLife 2020;9:e50135. DOI: https://doi.org/10.7554/eLife.50135 4 of 23
Research article Cell Biology
I. Epithelium + CRC Tumor Cells
II. Fibroblasts + Epithelium + CRC
Tumor Cells
A. B.
C. D.
Caco2 Alone Caco2 + HCT116
Epithelium
Endothelium
ZO-1 DAPI
VE-Cadherin E-Cadherin
HCT116 H2B GFP DAPI
0 2 4 6 8 10
0
1
2
3
Day
P
app
(cm/s) x 10
-6
Caco2 Alone
Caco2 + HCT116
9
Table 1.1. Chip locations specified for cell types and ECM composition used in the CRC-
Chip.
Channel Cell Type Cell Fluorescent Tag ECM
Bottom Endothelial HUVEC RFP Matrigel and
Collagen I
Top Epithelial Caco2 C2BBe1 Unlabeled Matrigel and
Collagen I
Top Stroma (CAF) 000UE, 000UK,
000U8, 000US
Cell Tracker Deep
Red
Collagen IV, Matrigel
and Collagen I
overlay
Top Cancer Cell
Line
HCT116 or HT29 H2B-GFP Matrigel and
Collagen I
Top CRC Organoid ORG000US H2B-GFP Matrigel
10
Figure 1.5. Tight junction formation and tumor cell morphology in the HCT116-CRC-Chip. A. Tiled maximum
projection (60 µm Z-height) and zoomed-in (white box) confocal fluorescent images of the epithelial channel of the
Caco2 Intestine Chip on day 6 stained for tight junction protein ZO-1 (gold). Cell nuclei are labeled with DAPI (blue).
Scale bar represents 1 mm on the tiled image and 200 µm on the zoomed-in image. B. Tiled maximum projection (60
µm Z-height) and zoomed-in (white box) confocal fluorescent images of the epithelial channel of the HCT116-CRC-
Chip on day 6 stained for tight junction protein ZO-1 (gold). HCT116 are labeled with H2B-GFP (green) and cell
nuclei are labeled with DAPI (blue). Scale bar represents 1 mm on the tiled image and 200 µm on the zoomed-in
image. C. HCT116 H2B-GFP morphology was compared between traditional 2D cell culture and on-chip. D. Tiled
confocal fluorescent image of the entire CRC-Chip. HCT116 tumor cells are H2B-GFP labeled, and HUVEC cells are
RFP labeled. The nuclei of the top epithelial channel were stained with DAPI.
HCT116 Traditional
Cell Culture
HCT116
CRC-on-Chip
A.
B.
Caco C2BBe1 HUVEC RFP HCT116 H2B GFP
C. D.
Caco2 Alone
ZO-1 DAPI
Caco2+HCT116
ZO-1 DAPI HCT116 H2B GFP
11
Metabolic Comparison of Intestine Chips versus CRC-Chips
The microfluidic nature of the CRC-Chip system supports dynamic measurements of the
effluent. To determine whether this CRC-Chip model mimics important aspects of CRC biology,
I performed mass spectrometry-based metabolomics. Metabolite extracts of inlet and outlet
media from the top epithelial channel and the bottom endothelial channel were analyzed from
Intestine Chips and diseased CRC-Chips on days 0 (D0) and 6 (D6). In order to better
understand the metabolomic profiles across different stages of CRC aggressiveness, I
performed experiments with diseased CRC-Chips seeded with HCT116 (HCT116-CRC-Chip)
and HT29 (HT29-CRC-Chip) cell lines (Tables 1.2-1.5). A principal component analysis (PCA)
of the metabolite intensities shows clear separation between the epithelial and the endothelial
channel effluents for both the Intestine Chips and the CRC-Chips (Figure 1.6A). When
evaluating the metabolites in the epithelial channel, there was significant overlap in the Intestine
Chip between D0 and D6, resulting in only a few differential metabolites, while there were
several differential metabolites detected in the HT29-CRC-Chip, and a much larger number of
differential metabolites detected in the HCT116-CRC-Chip between days (Figure 1.6B).
Furthermore, metabolic profiles between the Intestine Chips and the HCT116-CRC-Chip on D6
yielded differentially expressed metabolites (Table 1.6). The differential metabolites that were
identified using our in-house library between the Intestine Chip and the HCT116-CRC-Chip
were mapped to pathways using Ingenuity Pathway Analysis (IPA) to identify the most affected
pathways in the HCT116-CRC-Chips (Table 1.7). The TCA cycle and several amino acid
metabolism pathways were the most significantly altered pathways, as highlighted in Figure
1.6C.
Separate clusters were observed between the timepoints in the HCT116-CRC-Chip
dataset (Figure 1.6B), with the identification of 50 significantly differentially expressed
metabolites in the epithelial effluent indicating CRC tumor cell growth. Differential metabolites
12
(p<0.05 and with a fold change greater than 2) in the epithelial effluent of the HCT116-CRC-
Chip mapped to ‘colorectal cancer’ disease state with the highest significance using IPA
(p=7.19E-12), while those from the HT29-CRC-Chip (p<0.05 and with a fold change greater
than 1) mapped to CRC to a less significant degree (p=2.31E-09). To confirm the specificity of
this finding, I performed 20 permutations in which I selected 50 random metabolites from our
library of identified compounds from the HCT116-CRC-Chip experiments and these “random
sets” were evaluated with the same pathway analysis. The mapping of our HCT116-CRC-Chip
data to ‘colorectal cancer’ and other related disease states is not based on chance, suggesting
that our CRC-Chip is a good model system to further study CRC progression (Figure 1.6D).
13
Figure 1.6. Metabolic analyses validate use of CRC-Chip to model CRC progression. (A) Epithelial and
endothelial effluent was collected from the Intestine Chip and the HT29- and HCT116-CRC-Chips on days 0 and 6 of
the experiment and mass spectrometry-based metabolomics was performed. The principal component analysis
(PCA) on the differential metabolites demonstrates the clustering of samples corresponding to the effluent
compartment (epithelium or endothelium) and the different time points. (B) Volcano plots comparing the metabolites
from the top epithelial channel on day 0 and day 6 for the Intestine Chip and the HT29- and HCT116-CRC-Chips.
Each point represents a metabolite. Analytes with p-values <0.05 and fold change >2 were regarded as statistically
significant (colored red and blue upregulated and downregulated, respectively). (C) Differential metabolites between
the Intestine Chip and the HCT116-CRC-Chip from the epithelial effluent showed altered TCA cycle and amino acid
metabolism via Ingenuity Pathway Analysis (IPA). (D) The 50 differential metabolites that matched to our internal
database from the epithelial channel of the HCT116-CRC-Chip mapped to colorectal cancer with the highest
significance (highest –Log10p-value) using IPA. Each group is ranked by p-value and colored based on the 50
differentially expressed metabolites from our data set, termed “CRC-Chip” (red; asterisk denotes significant outliers)
or 20 permutations of 50 randomly selected metabolites, termed “random set” (black box plots).Wilcoxon signed rank
test compared the CRC-Chips to the random selection sets for the ‘colorectal cancer’ disease state, p = 0.0005.
A.
C.
Intestine Chip Effluent HCT116-CRC-on-Chip Effluent
D0D6 Epithelial Channel D0D6 Endothelial Channel
D.
Non-Diseased Chip
Epithelial Effluent
HT29-CRC-on-Chip
Epithelial Effluent
HCT116-CRC-on-Chip
Epithelial Effluent
HT29-CRC-on-Chip Effluent
Component 1 (59.32%)
Component 2 (10.65%)
Component 1 (62.78%)
Component 2 (13.89%)
Component 1 (67.76%)
Component 2 (14.51%)
B.
14
Table 1.2. Differential metabolites between D0 and D6 of epithelial channel effluent for the
HCT116-CRC-Chip. Related to Figure 1.6.
Database
source
Metabolite Presumptive ID Observed
Molecular
Weight
Fold
Change
p value
Metabolites
matched to
in-house
IROA library
Citramalate 148.0374 -2.9 1.05E-03
5-Oxo-l-proline 129.0431 -3.46 1.45E-05
Threonine 119.0584 -2.16 2.71E-03
Pyruvate 88.0158 -3.39 2.04E-02
Proline 115.0634 -3.33 2.27E-02
Methionine 149.0509 7.1 2.70E-03
6-Phosphogluconic Acid 322.022 9.62 4.20E-02
Methyl Acetoacetate 116.0468 -2.73 3.02E-03
3-Dehydroshikimate 172.0315 2.97 7.31E-04
Pyridoxal 166.0854 -2.17 2.25E-02
Indole-3-Pyruvic Acid 263.077 -2.2 2.39E-03
Kynurenine 268.1034 -2.25 2.97E-03
Pyrrole-2-Carboxylate 171.051 -2.8 2.34E-02
2'-Deoxyuridine 5'-Monophosphate 308.0401 -2.12 4.26E-09
Pantolactone 130.0622 -3.09 1.75E-05
Inosine 268.0808 10.2 7.12E-05
Orotate 216.0387 -3.3 2.24E-02
Citrate 192.0271 -4.25 4.99E-04
Glycine 75.0322 -4.32 2.14E-03
Homocystine 314.0578 -2.03 1.41E-03
Creatinine 113.0588 -3.34 1.92E-03
Pyruvic Aldehyde 72.0201 -2.23 1.33E-02
Uridine 244.0699 -3.76 1.03E-02
Ascorbate 88.0161 4.57 4.99E-04
Tryptophan 204.0875 3.19 2.85E-03
Cysteine 121.0198 -3.32 4.99E-04
Adenosine 3',5'-Cyclic Monophosphate 375.0513 -3.43 1.15E-04
Folic Acid 441.1407 -7.3 4.20E-02
Glycerate 106.0262 -2.61 1.67E-02
Glutamine 146.0699 6.16 1.09E-04
Dihydrofolate 503.1686 1,460 2.29E-03
5-Phospho-D-Ribose 1-Diphosphate 435.9554 -4.05 3.60E-05
Malate 134.0213 -2.27 8.78E-04
2,6-Dihydroxypyridine 155.9546 -2.22 1.15E-04
N-Acetylneuraminate 309.1064 2.16 1.73E-03
3-(2-Hydroxyphenyl)Propanoate 212.0648 -2.6 2.58E-04
Tyrosine 181.0746 -3.11 2.85E-04
O-Acetyl-l-Serine 193.0535 2.43 2.28E-02
2-Methylmaleate 190.0469 -2.02 4.99E-04
15
1-Hydroxy-2-Naphthoate 188.0476 -4.04 7.20E-06
Lactate 90.0309 2.86 1.94E-05
2-Hydroxybutyric Acid 104.0451 3.11 4.26E-09
Lysine 146.1057 -2.1 9.82E-06
Taurine 125.0148 2.55 3.03E-03
Norleucine 131.091 2.22 2.06E-02
3-Methyl-2-Oxindole 193.0739 -1.61 4.52E-06
Cystine 240.0238 6.06 1.73E-03
Urate 168.0291 -2.54 6.80E-05
Alpha-Ketoglutaric Acid 146.0216 -4.15 4.16E-05
Dehydroascorbate 174.0173 -2.33 1.72E-03
Xanthine 152.033 -5.99 2.39E-02
Top
metabolites
matched to
METLIN
library or
Metabolites
with
molecular
formula
generated
(+)-Chebulic acid 402.0418 -11.1 2.89E-06
1-(Malonylamino)cyclopropanecarboxylic
acid
187.0461 2.42 4.24E-05
1,8-Naphthyridine-3-carboxylic acid, 1-
ethyl-1,4-dihydro-7-hydroxy-4-oxo-
280.068 -2.21 9.23E-03
11-O-Demethylpradinone I 510.0745 -2.39 1.67E-03
2-(1,2,3,4-Tetrahydroxybutyl)-6-(2,3,4-
trihydroxybutyl)pyrazine
304.1316 2.81 7.56E-03
2-(beta-D-Glucosyl)-sn-glycerol 300.1057 -7.34 1.35E-06
2,7-Anhydro-alpha-N-acetylneuraminic
acid
291.0952 -2.78 2.84E-03
2-Aminoadenosine 282.1084 -2.04 5.61E-04
4,5-Dihydroxyphthalate 198.0143 2.05 2.21E-04
4-Carboxy-2-oxo-3-hexenedioate 248.0215 2.62 2.29E-03
5-aminosalicyluric acid 256.067 2.57 9.07E-03
5-Hydroxymethylsulfamethoxazole 269.0496 2.6 2.82E-05
5'-Methoxybilobetin 642.1323 2.52 5.72E-03
7-Hydroxyriluzole 296.0016 -4.04 2.18E-03
Acromelic acid A 310.0757 2.15 3.73E-03
Ala His Cys 329.1148 2.52 1.73E-03
Asn His Gly 326.1328 3.33 4.49E-03
Asp Gly Ser 276.062 -2.69 4.69E-04
C12 H23 N O16 437.1014 2.98 6.30E-05
C16 H31 N22 531.3097 2.17 3.28E-03
C22 H21 N3 O21 663.0698 -2.3 1.18E-06
C24 H21 N6 O22 745.0721 -2.37 1.57E-06
C26 H21 N9 O23 827.0735 -2.3 4.20E-05
C27 H13 N9 O 479.1214 -2.1 5.04E-03
C35 H28 N19 O 730.2743 2.32 1.68E-03
C36 H21 N11 O24 991.0644 -2.16 4.47E-04
C40 H17 N10 O17 909.0706 -2.23 1.80E-04
C5 H4 N2 92.0362 -2.99 1.69E-04
C6 H4 N O10 249.9826 2.72 4.41E-03
16
C7 H10 O6 S 222.0191 -4.88 1.66E-05
C9 H2 N3 S2 215.9685 2.93 1.75E-03
CAY10583 433.1924 2.37 2.00E-04
Ceforanide 519.1029 2.85 2.92E-04
Chloroneb 205.008 -2.41 1.37E-03
Coriose 270.0968 -2.49 6.75E-03
Cysteinyl-Cysteine 224.0284 2.16 5.23E-03
Daidzein 7-O-glucuronide 429.1033 2.01 1.40E-04
De-O-methylsimmondsin 361.1378 6.46 5.87E-04
Dihydropteroic acid 360.1175 -2.33 2.97E-04
His Ala Ala 297.1433 4.5 7.42E-04
Hydroxymethylphosphonate 157.9987 2.05 6.31E-04
Ibudilast 230.1417 -7.81 2.53E-06
Inositol cyclic phosphate 242.021 2.29 2.76E-03
Leucodelphinidin 3-[galactosyl-(1-4)-
glucoside]
646.1776 2.42 7.83E-03
Leu-Nap-OH 482.1729 3.31 5.49E-03
Met Met Glu 455.139 2.02 6.32E-03
N6-Carbamoyl-L-threonyladenosine 458.1429 -2.49 1.62E-03
Ornaline 308.1198 -11.1 1.85E-05
Oryzalin 346.0933 -6.05 4.01E-04
Phenyl sulfate 173.9986 -6.1 1.57E-04
Saphenic acid methyl ester 282.1021 -2.51 9.40E-03
Thio-THIP 202.0418 -4.01 4.14E-03
Thr-asn-OH 355.0981 3.25 1.46E-05
Triamcinolone acetonide sulfate 514.1612 5.14 8.44E-03
UDP-L-Ara4N 581.0669 -2.18 2.55E-06
Wybutoxine 468.1618 2.33 4.05E-03
Yersiniabactin 527.1084 2.15 6.38E-05
PC(O-16:0/2:0)[U] 745.5598 11.1 1.19E-07
PC(16:0/18:1(9Z))[S] 731.544 -3.17 4.26E-09
PC(O-16:0/2:0)[U] 523.3626 11.1 9.54E-06
PC(18:1(9Z)/0:0) 507.3684 7.54 2.15E-03
PC(18:2(2E,4E)/18:2(2E,4E)) 781.5593 -2.36 2.09E-09
PE(22:4(7Z,10Z,13Z,16Z)/19:1(9Z)) 807.5744 2.87 4.60E-07
PE(18:3(9Z,12Z,15Z)/21:0) 783.5747 2.69 4.42E-10
PE(14:0/21:0)[U] 733.56 2.29 1.15E-09
PC(16:0/16:1(9Z)) 731.544 2.6 2.10E-05
PC(15:0/20:4(5Z,8Z,11Z,14Z)) 745.5598 2.84 2.20E-07
PC(O-16:0/3:0) 537.378 1250 3.84E-10
PC(O-16:0/O-2:0) 509.3829 751 9.44E-11
PC(O-16:1(9Z)/0:0) 479.3367 130 3.20E-05
PC(O-16:0/0:0)[U] 481.3522 887 2.86E-05
PE(22:4(7Z,10Z,13Z,16Z)/19:1(9Z)) 807.5744 2.87 4.62E-07
17
PC(O-18:1(9Z)/0:0) 549.3785 723 4.09E-13
PC(7:0/O-8:0) 481.3164 484 2.69E-11
PA(8:0/8:0) 446.2067 2.04 1.57E-08
PC(13:0/17:0) 705.528 11.3 5.84E-03
Table 1.3. Differential metabolites between D0 and D6 of endothelial channel effluent for
the HCT116-CRC-Chip, Related to Figure 1.6.
Database
source/Library
Metabolite Presumptive ID Observed
Molecular
Weight
Fold
Change
p value
IROA 5-Phospho-D-Ribose 1-Diphosphate 435.9554 5.07 2.58E-02
IROA 3-(4-Hydroxyphenyl)Lactate 182.0583 3.33 2.84E-02
IROA Alpha-Ketoglutaric Acid 146.0216 3.21 2.79E-02
IROA Citramalate 148.0374 2.68 9.39E-06
IROA O-Acetyl-l-Serine 193.0535 2.98 2.73E-02
IROA Itaconate 130.0264 2.94 2.57E-02
IROA Tryptophan 204.0874 2.58 1.73E-02
IROA Pyruvate 88.0158 -3.26 3.89E-03
IROA Homocystine 314.0578 -3.26 4.58E-05
IROA 1-Hydroxy-2-Naphthoate 188.0476 -2.28 1.54E-02
IROA Methyl Acetoacetate 116.0468 -4.39 1.13E-02
IROA 3-(2-Hydroxyphenyl)Propanoate 212.0648 -3.49 2.97E-02
IROA Normetanephrine 243.1021 -3.54 1.23E-02
IROA Dehydroascorbate 174.0173 -3.63 2.77E-02
IROA Malate 133.0377 -2.16 3.81E-02
IROA 2,6-Dihydroxypyridine 155.9546 -2.52 3.35E-05
IROA Lactate 90.0309 2.01 3.08E-02
IROA Proline 115.0634 -2.01 1.21E-02
IROA N-Acetyl-dl-Glutamic Acid 189.1651 -2.63 3.67E-02
IROA N-Acetyl-l-Aspartic Acid 175.1396 -4.62 2.87E-02
IROA Inosine 268.0808 2.39 1.27E-02
IROA Ferulate 194.1806 -9,680 3.50E-03
IROA Butanal 132.0769 -2.19 3.52E-02
IROA Sarcosine 89.0477 -2.06 4.66E-03
IROA Uridine-5-Monophosphate 324.1821 -31 2.20E-03
METLIN
(+)-Chebulic acid 402.0418 -2.67 4.77E-02
METLIN 2,7-Anhydro-alpha-N-
acetylneuraminic acid
291.0952 -3.45 1.76E-03
METLIN 2-Aminoadenosine 282.1084 -2.85 8.90E-04
METLIN 3'-Amino-3'-deoxy-AMP 392.0786 -4.61 3.98E-02
METLIN 3'-O-Methyl-(-)-epicatechin-7-O-
sulphate
414.0609 -3.63 2.33E-02
METLIN 5'-Methoxybilobetin 642.1323 -2.73 1.97E-02
METLIN 8-Hydroxy-3-chlorodibenzofuran 218.0109 3.35 3.26E-02
18
METLIN AVE-1625 556.0857 -2.79 1.90E-02
Molecular
Formula
C19 H42 N2 O22 650.2241 3.49 1.75E-02
Molecular
Formula
C22 H21 N3 O21 663.0698 -4.71 2.40E-05
Molecular
Formula
C24 H21 N6 O22 745.0721 -2.75 4.43E-05
Molecular
Formula
C26 H21 N9 O23 827.0735 -2.73 2.02E-05
Molecular
Formula
C28 H29 O15 605.1506 -2.2 4.18E-02
Molecular
Formula
C4 H2 N O3 S2 175.9456 2.39 2.34E-02
Molecular
Formula
C40 H17 N10 O17 909.0706 -3.64 2.46E-04
Molecular
Formula
C5 H4 N2 92.0362 -2.07 1.07E-02
METLIN Dihydropteroic acid 360.1175 -2.78 3.08E-02
METLIN Distemonanthin 404.0397 -2.7 4.19E-02
METLIN fumarylacetic acid 158.0212 2.82 3.71E-02
METLIN Galactosylglycerol 314.1227 2.07 8.27E-03
METLIN Hydrouracil (Dihydrothymine) 128.0589 2.34 4.33E-03
METLIN N-Nitrosodiethylamine 102.0796 2.3 7.07E-03
METLIN Phosphophosphinate 335.0578 -2.19 1.71E-04
Table 1.4. Differential metabolites between D0 and D6 of epithelial channel effluent for the
HT29-CRC-Chip, Related to Figure 1.6.
Database
source/Library
Metabolite Presumptive ID Observed
Molecular
Weight
Fold
Change
p value
IROA 1-Methyladenosine 281.1137
-12
1.14E-02
IROA 3-Ureidopropionate 132.0533
-2.91
7.45E-08
IROA 5'-Methylthioadenosine 314.1091
2.18
5.28E-03
IROA Raffinose 564.172
3.52
2.24E-03
IROA Aspartate 133.0373
3.31
2.96E-03
IROA Ornithine 132.0895
-2.43
9.24E-07
IROA Flavin Adenine Dinucleotide 785.1841
2.41
1.23E-03
IROA Inosine 268.0791
2.78
3.59E-03
IROA L-Proline 115.0629
-2.80
2.25E-09
IROA N-Acetylneuraminate 326.1354
6.45
8.41E-04
IROA Pyrrole-2-Carboxylate 133.011
-3.52
2.44E-12
IROA Sarcosine 89.0471
-3.31
3.72E-10
IROA Sucrose 342.1176
-2.60
1.03E-03
IROA Xanthine 152.0325
-3.80
5.62E-06
IROA Citrate 192.0271 -2.17 2.61E-04
19
IROA 5-Oxo-l-proline 129.0431 -2.70 1.57E-04
IROA Lactate 90.0312 -2.4 5.09E-03
IROA
Ferulate 194.0573 -6.16 1.04E-03
IROA 1-Methyl-6,7-Dihydroxy-1,2,3,4-
Tetrahydroisoquinoline 179.0941 -3.75 3.28E-02
IROA
L-Kynurenine 208.084 -23 1.54E-02
IROA
Cholesteryl Acetate 428.365 2.28 4.50E-02
IROA
2-Hydroxybutyric Acid 126.0315 -53 9.31E-04
METLIN
1,2-Dihydroxynaphthalene-6-
sulfonate 286.0166 3.11 2.00E-03
METLIN 2-(beta-D-Glucosyl)-sn-glycerol
300.103 -3.1 7.82E-04
METLIN 2-C-Methyl-D-erythritol 2,4-
cyclodiphosphate 323.9953 -2.2 4.32E-04
METLIN 2-Phenylaminoadenosine
358.1384 -4.1 3.73E-05
METLIN 3-Hydroxy-L-tyrosyl-AMP
526.1335 2.51 1.57E-04
METLIN 3-Hydroxy-OPC4-CoA
1003.2531 2.76 7.31E-04
METLIN 6''-(4-Carboxy-3-hydroxy-3-
methylbutanoyl)hyperin 608.1362 2.31 1.66E-04
METLIN 6-Mercaptopurine ribonucleoside 5'-
diphosphate 461.0225 -3.2 3.86E-11
METLIN AM679
417.0581 -3 1.54E-09
METLIN Azamethiphos
345.9607 2.16 2.50E-02
METLIN Butoconazole
456.0233 3.93 5.22E-03
Molecular
Formula
C13 H6 N O7
288.0157 3.01 2.41E-03
Molecular
Formula
C16 H11 N6 O16
543.0254 -3.4 6.42E-11
Molecular
Formula
C20 H13 N6 O18
625.0279 -3.2 4.80E-11
Molecular
Formula
C21 H15 N10 O16
663.065 -3.6 2.88E-08
Molecular
Formula
C21 H35 N8 O17 S
703.1831 2.62 2.37E-03
Molecular
Formula
C22 H17 N6 O20
685.0471 -3.3 2.54E-11
Molecular
Formula
C25 H15 N11 O18 S
789.0297 -2.6 2.82E-10
Molecular
Formula
C25 H32 N4 O24
772.1405 2.05 2.04E-04
Molecular
Formula
C27 H17 N19 O
623.1863 2.63 9.35E-04
Molecular
Formula
C27 H24 N18 O11
776.1849 3.1 1.96E-04
Molecular
Formula
C29 H29 N22 O4 S2
813.2188 2.72 1.10E-03
Molecular
Formula
C33 H54 N O26 S2
944.2348 4.5 3.85E-04
20
Molecular
Formula
C34 H13 N13 O14
827.0675 -2.6 3.79E-08
Molecular
Formula
C34 H13 N14 O21
953.0244 -2.3 6.97E-09
Molecular
Formula
C34 H15 N8 O21
871.0308 -2.4 1.82E-09
Molecular
Formula
C35 H11 N11 O10
745.0668 -3.4 4.98E-08
Molecular
Formula
C42 H19 N13
705.187 2.34 9.18E-04
METLIN Cyanidin 3-O-[b-D-Xylopyranosyl-(1-
2)-[(4-hydroxybenzoyl)-(-6)-b-D-
glucopyranosyl-(1-6)]-b-D-
galactopyranoside] 862.2344 4.4 2.04E-04
METLIN Eujambolin
542.1054 2.39 3.05E-03
METLIN Floxacillin
499.0614 -3.1 1.53E-08
METLIN Flupyrsulfuron-methyl
465.0653 -2.1 4.78E-09
METLIN Gossypetin 3-sophoroside-8-
glucoside 821.206 3.78 2.22E-03
METLIN His Phe Gly
381.1381 34.1 2.05E-03
METLIN Isoscutellarein 7-(6'''-acetylallosyl-(1-
2)-6''-acetylglucoside) 694.1832 3.22 2.54E-04
METLIN Leucodelphinidin 3-[galactosyl-(1-4)-
glucoside] 646.1742 3.13 6.06E-03
METLIN Leu-Nap-OH
482.1693 3.59 3.80E-03
METLIN L-Rhamnulose 1-phosphate
304.0556 4.83 4.93E-03
METLIN N-(4-
hydroxyphenyl)ethoxycarbothioamide 197.0535 -2.6 2.90E-06
METLIN N-benzyl-1-methyl-1H-pyrazolo[3,4-
d]pyrimidin-4-amine 239.117 -2 2.23E-02
METLIN N-Cyclohexylformamide
127.1005 -1100 4.94E-07
METLIN Obtusol
472.0021 2.6 1.78E-03
METLIN Perfluidone
379.0197 -3 2.66E-12
METLIN Phosphophosphinate
297.0171 -3.8 1.80E-10
METLIN Photinus luciferin
325.9949 -2.3 2.47E-04
METLIN Primflaside
728.1765 3 5.92E-04
METLIN PtdIns-(4,5)-P2 (1,2-dihexanoyl)
690.1374 2.2 1.67E-04
METLIN Pyridoxamine-5'-Phosphate
294.0578 -180 1.90E-04
METLIN Tamarixetin 3-O-sulfate
396.0145 3.51 1.85E-03
METLIN
Thioridazine 2,5-disulfoxide 419.1748 -2.7 3.87E-05
METLIN
UDP-L-Ara4N 581.0635 -3.5 7.06E-09
METLIN
PC(O-12:0/2:0) 467.3018 15 4.22E-02
METLIN
L-Arogenate 227.0789 -13 4.47E-02
METLIN
PS(P-16:0/15:1(9Z)) 703.4809 2.32 4.54E-02
METLIN
Glu Tyr 310.1154 -3.22 3.88E-02
21
Table 1.5. Differential metabolites between D0 and D6 of endothelial channel effluent for
the HT29-CRC-Chip, Related to Figure 1.6.
Database
source/Library
Metabolite Presumptive ID Observed
Molecular
Weight
Fold
Change
p value
IROA N-Acetylneuraminate 326.1354 2.64 2.79E-03
IROA Sucrose 342.1176 -53 2.02E-03
IROA
Hypoxanthine 136.038 3.62 1.60E-03
IROA
2-Hydroxybutyric Acid 126.0315 -2.31 2.44E-02
IROA Inosine 268.0791 2.26 2.00E-03
METLIN His Phe Gly 381.1381 2.15 3.74E-03
METLIN N-benzyl-1-methyl-1H-pyrazolo[3,4-
d]pyrimidin-4-amine
239.117 -4.1 1.06E-04
METLIN N-Cyclohexylformamide 127.1005 -14 5.96E-03
METLIN 3,4-Dihydroxyphenylglycol O-sulfate 250.0127 -2 4.50E-02
METLIN Thr Thr Glu 366.1749 -53 3.86E-02
METLIN L-Rhamnulose 1-phosphate 304.0556 2.2 1.90E-03
METLIN Suprofen 306.054 2.07 1.71E-03
METLIN 2-Phenylaminoadenosine 358.1384 -17 2.11E-03
METLIN
Nicarbazin 302.0626 -5.76 4.20E-03
METLIN 2-Hexaprenyl-6-methoxy-1,4-
benzoquinol 565.447 21 3.66E-02
METLIN
His Ala Ser 313.138 6.99 4.62E-02
Molecular Formula
C25 H8 N8 O18 707.9943 67 2.98E-02
Table 1.6. Differential metabolites between the Intestine Chip and the HCT116-CRC-Chip epithelial
effluent on D6, Related to Figure 1.6.
Database
Source/Library
Metabolite Presumptive ID Observed
Molecular
Weight
Fold
Change
p value
IROA
1-Hydroxy-2-Naphthoic acid 188.0476 2.49 5.96E-06
IROA Alpha-Ketoglutaric Acid 146.0216 2.14 1.81E-03
IROA 3-Dehydroshikimic acid 172.0315 -2.64 1.16E-02
IROA 3-Hydroxyphenylacetic acid 182.0583 2.15 4.90E-02
IROA 3-Methyloxindole 193.0739 2.01 3.76E-02
IROA Ascorbate 88.0161 4.58 4.43E-04
IROA Citrate 192.0271 1,410 3.28E-03
IROA Galactaric acid 210.0375 96.3 3.74E-02
IROA Dihydrofolate 503.1686 -1,060 2.36E-02
IROA Folic acid 441.1407 1.21 2.85E-04
IROA Guanine 151.0495 -1.99 1.57E-02
IROA Guanosine 283.092 -1.57 4.02E-02
IROA Inosine 268.0808 -1.33 1.55E-02
IROA Alanine 89.0471 1.94 2.65E-02
22
IROA Cysteine 121.0198 -1.8 2.47E-03
IROA Cystine 240.0246 -1.23 3.16E-02
IROA Glutamic acid 147.1306 -1.28 1.20E-02
IROA Glutamine 146.0699 -1.02 8.66E-03
IROA Homocysteine 314.0578 1.48 1.60E-03
IROA Lactate 90.0309 1.83 2.22E-04
IROA Malate 134.0213 5.16 3.46E-02
IROA Methionine 149.0509 -1.13 7.94E-02
IROA Normetanephrine 243.1021 3.04 3.28E-05
IROA Maleamate 115.0263 -4.66 9.06E-06
IROA 3-Ureidopropionic acid 132.0536 6.62 2.50E-05
IROA O-Acetyl-L-serine 193.0535 -1.06 2.73E-02
IROA Oxalacetate 132.0064 1.27 3.60E-02
IROA 5-Oxo-l-proline 129.0431 2.7 1.76E-02
IROA Pyruvate 88.0158 -3.82 7.74E-06
IROA Succinate 118.0258 1.01 2.60E-02
IROA Urate 168.0291 2.73 2.14E-02
IROA Xanthine 152.033 2.63 2.82E-04
IROA Xanthosine 284.0749 2.01 7.34E-03
IROA Itaconate 130.0264 -6.26 3.21E-02
METLIN
(±)12-HETE 320.2315 -3.83 2.04E-02
METLIN (3R,7R)-1,3,7-Octanetriol 162.1248 2.29 1.18E-02
METLIN (S)-a-Amino-2,5-dihydro-5-oxo-4-
isoxazolepropanoic acid N2-glucoside
334.0951 -3.14 4.08E-02
METLIN 1-(6-[3]-ladderane-hexanoyl)-2-(8-[3]-
ladderane-octanyl)-sn-
glycerophosphocholine
765.5646 -4.3 5.16E-04
METLIN 1,15-Hexadecadien-3-one 238.229 -2.71 5.33E-03
METLIN 1,3,8-Naphthalenertriol 176.0469 2.81 5.98E-03
METLIN 12-hydroxy-10-dodecenoic acid 214.1566 7.49 1.02E-03
METLIN 13E-Docosenamide 337.3323 3.05 3.05E-02
METLIN 1-alpha-Acevaltrate 480.203 -2.44 4.96E-03
METLIN 1-Palmitoyllysophosphatidylcholine 495.3306 2.41 1.65E-02
METLIN 2,2'-(3-methylcyclohexane-1,1-
diyl)diacetic acid
214.12 -4.54 4.07E-05
METLIN 2,3-Dihydroxy-2,4-cyclopentadien-1-one 112.0134 2.17 1.82E-02
METLIN 2-Acetylpyrrolidine 113.0836 -2.52 2.23E-03
METLIN 2E-Decenedioic acid 217.1306 -6.14 7.44E-04
METLIN 2-Hydroxyiminodibenzyl glucuronide 409.1129 -2.74 3.10E-02
METLIN 3-(10-Heptadecenyl)phenol 347.3181 3.29 2.29E-02
METLIN 3-Deoxyvitamin D3 368.3437 2.69 1.97E-02
METLIN 3-methoxy Limaprost 429.3084 -2.55 3.05E-02
METLIN 3Z,6Z,9Z,12Z,15Z-Pentacosapentaene 342.3322 3.78 4.29E-02
23
METLIN Arg Arg Gln 480.2531 -4.71 1.01E-04
METLIN Asp-Gly-OH 298.0388 2.59 1.45E-03
Molecular Formula C10 H22 O4 206.1514 2.31 1.42E-03
Molecular Formula C10 H24 N3 O4 250.1772 3.14 4.96E-03
Molecular Formula C11 H23 N4 O 227.1882 2.12 1.51E-03
Molecular Formula C12 H5 N4 O5 S2 348.9695 3.78 1.27E-02
Molecular Formula C15 H15 N14 O3 439.1406 -2.04 1.64E-02
Molecular Formula C16 H15 N13 S2 453.1012 2.22 2.12E-02
Molecular Formula C18 H38 N4 310.3055 2.69 1.29E-02
Molecular Formula C19 H18 N O15 S 532.0416 3.29 2.16E-02
Molecular Formula C21 H37 N 303.292 2.42 1.08E-03
Molecular Formula C22 H42 N16 530.3775 2.91 1.12E-02
Molecular Formula C27 H19 N4 O26 815.032 2.75 4.56E-02
Molecular Formula C29 H20 N15 O19 882.1052 -3.71 1.07E-03
Molecular Formula C41 H24 N2 O11 720.1393 2.43 4.63E-02
Molecular Formula C50 H81 N10 821.6635 -2.56 1.72E-02
Molecular Formula C6 H14 O4 150.0883 2.1 3.59E-02
METLIN Cer(d16:1/24:0) 637.5976 2.93 4.28E-02
METLIN Cer(d16:1/24:0) Esi+16.052986 638.621 2.04 3.98E-02
METLIN Cys Glu Asp 382.1085 3.53 1.28E-02
METLIN Dioctyl hexanedioate 370.3076 -2.86 1.38E-02
METLIN GlcCer(d18:2/23:0) 812.6762 3.23 3.02E-03
METLIN Glycerol tripropanoate 282.1076 -3.11 0.00E+00
METLIN Kojic Acid 142.0242 2.23 2.45E-03
METLIN LacCer(d14:1/16:0) 805.5577 3.8 4.16E-02
METLIN LysoPC(16:1(9Z)) 493.3156 -2.6 1.85E-02
METLIN LysoPC(20:4(8Z,11Z,14Z,17Z)) 543.3304 2.41 3.52E-02
METLIN MG(16:0/0:0/0:0)[rac] 330.2761 16 0.00E+00
METLIN Nap-Thr-OH 424.13 -2.22 2.62E-02
METLIN PA(P-20:0/22:0) 772.6457 -5.55 2.37E-02
METLIN PC(13:0/17:0) 705.528 -3.18 3.73E-04
METLIN PC(15:0/20:4(5Z,8Z,11Z,14Z)) 745.5598 -2.79 8.66E-03
METLIN PC(16:0/18:1(9Z))[S] 759.5736 -3.94 3.01E-04
METLIN PC(16:1(9Z)/22:2(13Z,16Z)) 811.6049 3.46 2.79E-02
METLIN PC(18:0/18:1(6Z)) 787.6068 2.94 2.47E-02
METLIN PC(18:0/20:4(8Z,11Z,14Z,17Z)) 809.5885 -2.89 3.30E-04
METLIN PC(18:1(17Z)/18:1(17Z)) 785.5891 2.15 4.84E-02
METLIN PC(18:1(9Z)/0:0) 521.3467 -2.62 1.33E-03
METLIN PC(18:1(9Z)/22:4(7Z,10Z,13Z,16Z)) 835.6032 -2.82 5.71E-04
METLIN PC(18:2(2E,4E)/18:2(2E,4E)) 781.5593 -2.66 2.66E-03
METLIN PC(19:1(9Z)/15:0) 759.5749 -3.63 4.50E-04
24
METLIN PC(7:0/O-8:0) 481.3164 2.21 7.33E-03
METLIN PC(O-16:0/3:0) 537.378 -3.94 1.14E-02
METLIN PC(O-16:0/O-2:0) 509.3829 -2.72 2.30E-02
METLIN PC(O-16:1(9Z)/0:0) 479.3367 2.17 3.72E-02
METLIN PC(O-18:1(9Z)/0:0) 507.3684 3.19 2.65E-02
METLIN PE(14:0/21:0)[U] 733.56 2.2 4.29E-03
METLIN PE(18:3(9Z,12Z,15Z)/21:0) 783.5747 -5.55 2.30E-03
METLIN PE(22:4(7Z,10Z,13Z,16Z)/19:1(9Z)) 807.5744 -3.18 3.23E-02
METLIN PE(P-20:0/17:1(9Z)) 743.5794 -2.68 3.76E-02
METLIN Phe Gly Val 321.1685 3.25 1.34E-03
METLIN Phe His Ile 436.227 2.76 1.97E-02
METLIN Phe-Tyr-OH 453.1569 3.24 5.98E-03
METLIN PS(O-18:0/21:0) 836.6705 2.48 3.21E-02
METLIN Triphenylphosphine oxide 278.0856 3.95 7.80E-04
METLIN Triphyllin A 656.226 2.97 3.06E-03
METLIN UH-301 287.173 2.83 3.56E-02
METLIN Varenicline 228.1307 -3.29 1.35E-02
METLIN Vitamin E succinate(tocopherol
succinate)
530.3994 -2.69 3.43E-03
METLIN Ximelagatran 490.2893 4.12 1.78E-03
Table 1.7. Top significantly altered metabolic pathways from IPA analysis of differential
metabolites between Intestine Chip and HCT116-CRC-Chip epithelial effluent on D6. Related to
Figure 1.6.
Metabolic Pathway p value Metabolites
Alanine Metabolism 3.04E-04 2-oxoglutaric acid, L-alanine, L-glutamic acid, pyruvic acid
TCA Cycle II
(Eukaryotic)
2.95E-04 2-oxoglutaric acid, citric acid, fumaric acid, L-malic acid,
oxalacetic acid, succinic acid
Purine Nucleotides
Metabolism
8.84E-03 adenosine, guanine, guanosine, hypoxanthine, inosine, uric
acid, xanthine, xanthosine
Adenosine Nucleotides
Metabolism
2.15E-05 adenosine, hypoxanthine, inosine, uric acid, xanthine
Glutamate Metabolism 2.12E-04 2-oxoglutaric acid, fumaric acid, L-glutamic acid, oxalacetic
acid
Superpathway of
Methionine Degradation
8.42E-04 2-oxoglutaric acid, adenosine, L-cysteine, L-glutamic acid, L-
homocysteine, L-methionine, pyruvic acid
25
Examination of Tumor Cell Intravasation
When I examined the PCA for the endothelial channel effluent (Figure 1.6A), I observed
a similar pattern to what was detected in the epithelial channel. Separate clusters of metabolites
between D0 and D6 were identified in the HCT116-CRC-Chip, minimal separation was seen in
the HT29-CRC-Chip, and no separation was found in the Intestine Chip. I detected over 20
differential metabolites in the HCT116-CRC-Chip endothelial effluent and only 5 differential
metabolites in the HT29-CRC-Chip (Figure 1.7A), suggesting metabolic changes within the
endothelial channel occur over time only when CRC tumor cells are present in the model. When
I analyzed the differentially expressed metabolites from the endothelial compartment in the
HCT116- and HT29-CRC-Chips using IPA, I found that several pathways were significantly
altered in the HCT116-CRC-Chips (glycine betaine degradation, alanine metabolism, HIF1α
signaling, and adenine and adenosine salvage pathways) (Table 1.8) while no pathways were
identified as significantly altered in the HT29-CRC-Chips.
26
Figure 1.7. Validation of CRC tumor cell invasion from an epithelial to endothelial compartment, mimicking
intravasation. (A) Volcano plots comparing the metabolites from the bottom endothelial channel on day 0 and day 6
for the Intestine Chip and the HT29- and HCT116-CRC-Chips. Each point represents a metabolite. Analytes with p
values <0.05 and fold change >2 were regarded as statistically significant (colored red and blue). (B) 6 regions of the
chip were imaged via confocal microscopy and input into 3-D reconstruction software for GFP+ cell quantification. An
invasion ratio was calculated based on the number of GFP+ cells in the bottom channel compared to the top channel
and normalized by the day 0 counts. (C&D) Tumor cell (HCT116 or HT29) invasion was monitored over time by
imaging the same chip regions at various timepoints, days 0, 2, 6. Representative images show different invasion
behavior for each tumor cell (C) and quantification is also depicted (D). N=6 Chips. Data are represented as mean ±
SEM and analyzed using a 2-way ANOVA; *p<0.05; **p<0.01). (E) CRC organoids (H2B-GFP labeled) from patient
000US were dissociated and fragments were seeded onto the ECM coated epithelial channel. Brightfield image of the
000US organoids seeded in the epithelial channel of the chip showing the CRC tissue architecture on day 6 (D6).
Scale bars represent 100µm. Chip schematic courtesy of Emulate, Inc. (F) Invasion of ORG000US was measured on
D6 of the experiment (N=6 Chips). An invasion ratio was calculated based on the number of GFP+ cells in the bottom
channel compared to the top channel and normalized by the day 0 counts. Data are represented using a boxplot.
A. B.
E.
D0 D2 D6
HT29
HCT116
0 2 4 6
0
2
4
6
Days
Invasion Ratio Normalized to D0
HCT116
HT29
**
*
More Invasive
Less Invasive
D. C.
chip (Figure 1D), with well-defined epithelial tight junctions, as demonstrated by ZO-1 protein stain-
ing and endothelial adherent junctions visualized using antibodies against VE-cadherin
(Dawson et al., 2016). Importantly, these culture conditions resulted in a time-dependent improve-
ment of intestinal permeability as indicated by the low permeability coefficient (Papp) of fluores-
cently labeled dextran recorded in the Duodenum Intestine-Chip generated from organoid-derived
cells of three different individuals (Figure 1E). Overall, this data indicates that the human adult Duo-
denum Intestine-Chip supports the formation of a functional barrier with in vivo relevant cytoarchi-
tecture, cell-cell interactions, and permeability parameters.
To confirm differentiation of the organoid-derived cells within the chip into all of the distrinct epi-
thelial cell lineages as found in vivo, we assessed average mRNA gene expression levels of cell-type-
Figure 1. Duodenum Intestine-Chip: a microengineered model of the human duodenum. (a) Brightfield images of
human duodenal organoids (top) and human microvascular endothelial cells (bottom) acquired before their
seeding into epithelial and endothelial channels of the chip, respectively. (b) Schematic representation of
Duodenum Intestine-Chip, including its top view (left) and vertical section (right) showing: the epithelial (1; blue)
and vascular (2; pink) cell culture microchannels populated by intestinal epithelial cells (3) and endothelial cells (4),
respectively, and separated by a flexible, porous, ECM-coated PDMS membrane (5). (c) Scanning electron
micrograph showing complex intestinal epithelial tissue architecture achieved by duodenal epithelium grown for 8
days on the chip (top) in the presence of constant flow of media (30 ml/hr) and cyclic membrane deformations (10%
strain, 0.2 Hz). High magnification of the apical epithelial cell surface with densely packed intestinal microvilli
(bottom). See Figure 1-figure supplement demonstrating the effect of mechanical forces on the cytoarchitecture of
epithelial cells and the formation of intestinal microvilli (d) Composite tile scan fluorescence image 8 days post-
seeding (top) showing a fully confluent monolayer of organoid-derived intestinal epithelial cells (magenta, ZO-1
staining) lining the lumen of Duodenum Intestine-Chip and interfacing with microvascular endothelium (green, VE-
cadherin staining) seeded in the adjacent vascular channel. Higher magnification views of epithelial tight junctions
(bottom left) stained against ZO-1 (magenta) and endothelial adherence junctions visualized by VE-cadherin
(bottom right) staining. Cells nuclei are shown in gray. Scale bars, 1000 mm (top), 100 mm (bottom) (e) Apparent
permeability values of Duodenum Intestine-Chips cultured in the presence of flow and stretch (30 ml/hr; 10% strain,
0.2 Hz) for up to 10 days. Papp values were calculated from the diffusion of 3 kDa Dextran from the luminal to the
vascular channel. Data represent three independent experiments performed with three different chips/donor, total
of three donors; Error bars indicate s.e.m.
The online version of this article includes the following figure supplement(s) for figure 1:
Figure supplement 1. Flow-induced increase in primary intestinal epithelial cells height and microvilli formation.
Kasendra et al. eLife 2020;9:e50135. DOI: https://doi.org/10.7554/eLife.50135 4 of 23
Research article Cell Biology
ORG000US
0
5
10
15
20
CRC Organoid
D6 Invasion Ratio
(Normalized to D0)
F.
HCT116-CRC-on-Chip
Endothelial Effluent
Non-Diseased Chip
Endothelial Effluent
HT29-CRC-on-Chip
Endothelial Effluent
27
Table 1.8. Top significantly altered metabolic pathways from IPA analysis of differential
metabolites between Intestine Chip and HCT116-CRC-Chip endothelial effluent on D6.
Related to Figure 1.7.
Metabolic Pathway p value Metabolites
Glycine Betaine
Degradation
1.06E-04 L-homocysteine, pyruvic acid, sarcosine
Alanine Metabolism 3.42E-04 2-oxoglutaric acid, pyruvic acid
H1F1 α signaling 8.47E-04 lactic acid, pyruvic acid
Adenine and
Adenosine Salvage
pathway
2.60E-04 inosine, phosphoribosyl pyrophosphate
Based on the metabolic changes observed within the endothelial compartment when
CRC tumor cells of various aggressiveness were present on-chip, I hypothesized that the CRC-
Chip could help address important questions related to tumor metastasis, in particular early
tumor cell dissemination. Specifically, tumor cells were monitored via confocal microscopy to
visualize and quantify the invasion of cells from the top epithelial channel into the bottom
endothelial channel through an ECM coating and a porous membrane. Given that the tumor
cells traversed the epithelium and ECM to invade through the porous membrane and into the
endothelium channel, I termed this an “invasion assay”. I calculated a ratio of the number of
invaded cells (tumor cells in the bottom channel) to the number of non-invaded cells (tumor cells
in the top channel) over time to determine an invasion rate (Figure 1.7B). Data was normalized
to D0 to account for any initial cell seeding variability between chips (Figure 1.8A). While the
epithelial Caco2 C2BBe1 cells showed minimal invasion (Figure 1.8A), the invasion rate differed
between CRC tumor cell lines, with the HT29 H2B-GFP cells having a significantly lower
invasion rate as compared with the HCT116 H2B-GFP (Figure 1.7C and Figure 1.7D).
Specifically, the average day 6 invasion ratio for all three cell lines was: HCT116 = 4.97; HT29 =
1.15; Caco2 C2BBe1 = 0.37. This finding mimics the aggressiveness of the tumors from which
28
these cells were derived and is supported by extensive in vitro and in vivo data in which
HCT116 cells are highly aggressive and known to develop liver metastases in vivo, while the
HT29 cells possess less efficient metastatic capabilities (Olejniczak et al., 2018). This trend of
the HCT116 cells invading more than the HT29 cells corroborates the metabolic changes seen
in the endothelial compartment in the CRC-Chips (discussed above), suggesting the differential
metabolites and altered metabolic pathways may be a result of CRC tumor cells invading into
the endothelium.
Traditionally, in vitro tumor cell invasion has been studied using transwell assays. I
performed modified transwell experiments by culturing HUVEC cells on the bottom of the
transwell membrane and co-culturing Caco2 C2BBe1 and CRC tumor cells on top of the
membrane to mimic the culturing of multiple cell types in the chip system. HT29 cells were less
invasive than HCT116 cells in the transwell system; however, when the number of invaded cells
is compared between transwell and on-chip experiments at a similar time point (day 2 for both
experiments), the on-chip invasion assay indicated less invasion (Figure 1.8B). While the pore
size and the cells seeded per culture area were the same between the two different model
systems, the presence of mechanical cyclic stretching and fluid flow in the chip could explain
these differences. This suggests that the chip system more closely models in vivo intravasation
as a rare event compared to traditional in vitro assays (Deryugina and Kiosses, 2017; Hapach et
al., 2019; Wyckoff et al., 2000).
29
Figure 1.8. Validation of CRC-Chip intravasation assay. (A) Tumor cell (HCT116 or HT29) invasion was
monitored over time by imaging the same chip regions at various timepoints, days 0, 2, 6, as described in Figure 3.
Caco2 C2BBe1 cells were stained with Cell Tracker Green (CTG) and invasion of the Caco2 cells was measured in
the Intestine Chip (N=3 Chips) on days 0, 2, and 6 as described in Figure 3. Raw invasion ratios were plotted as
boxplots with Tukey’s rule indicating an outlier (black box). In order to account for outliers, all data was normalized
back to D0 invasion ratios as shown in Figure 3D. (B) Tumor cell invasion on-chip (day 2) was compared to traditional
transwell invasion assays. HUVEC cells were seeded onto the bottom of the transwell membrane and tumor cells
were seeded on the top of an ECM coated fluoroblok transwell and the GFP+ cells were counted 48 hours later on
the bottom of the membrane. (N=6 Chips). Data are represented as boxplots and analyzed using an unpaired t-test;
***p<0.001; ****p<0.0001.
Previous studies have combined the organ-on-chip system with organoids derived from
healthy small and large intestine tissues and demonstrated more in vivo-like composition of
intestinal cell types (Kasendra et al., 2018), mucus physiology (Sontheimer-Phelps et al., 2020)
and drug response (Kasendra et al., 2020). I adapted these published methods to seed
organoids derived from human colon tumors in the top channel of the chips (without the
presence of the Caco2 C2BBe1 cells) (Figure 1.7E). I confirmed that tumor organoid cells were
capable of invading into the endothelial channel under the same fluid and mechanical forces as
applied in the Intestine Chip (Figure 1.7F).
CRC Cells Exhibit Phenotypic Heterogeneity During Intravasation
Given the differential invasive phenotypes of HT29 or HCT116 cells, I sought to identify
changes occurring during intravasation in our CRC-Chips. First, I investigated the expression of
epithelial or mesenchymal markers in CRC-Chips cultured with HT29 or HCT116 tumor cells,
30
which differ in their EMT programming (Pino et al., 2010). HT29 tumor cells in the top, epithelial
channel formed tight clusters that stained strongly for the adhesion molecule E-cadherin and
weakly for vimentin, suggesting these cells are epithelial-like (Figure 1.9, left panels). In
contrast, staining of HCT116 tumor cells showed decreased E-cadherin expression towards the
center of the tumor clusters and stronger vimentin expression at the edges of these clusters
(Figure 1.9, right panels), suggesting that the HCT116 tumor cells are losing epithelial features
and gaining a more mesenchymal-like phenotype.
Figure 1.9. HT29 and HCT116 cells show different expression of epithelial and mesenchymal markers.
Representative images of HT29 and HCT116 clusters in top channel stained with E-cadherin and vimentin. Strong
vimentin positive staining is observed on the periphery of HCT116 cell clusters compared to HT29 cells on day 10.
Scale bars represent 100 µm.
I further investigated HCT116 cells that invaded and adhered to the bottom, endothelial
channel and discovered heterogeneity in the expression of epithelial or mesenchymal markers.
Some tumor cells stained strongly for E-cadherin and were negative for vimentin, forming large
clusters bordered by HUVECs (Figure 1.10A, top panels). While in other regions of the same
chip, there were invaded HCT116 cells that did not express E-cadherin, but stained positive for
vimentin and showed a mesenchymal-like morphology (Figure 1.10A, bottom panels). When I
examined HCT116 cells cultured in traditional cell culture conditions (plastic, 2D), I found less E-
cadherin-positive and vimentin-positive cells (Figure 1.10B). When I quantitated the percentage
of cells identified as positive or negative for E-cadherin or vimentin expression, I discovered that
31
there was greater heterogeneity in the invaded HCT116 cells on-chip compared to plastic
(Figure 1.10C). The percentage of E-cadherin+, E-cadherin+ and vimentin+, and vimentin+
HCT116 cells increased on-chip, while the percentage of cells that were negative for both E-
cadherin and vimentin decreased on-chip as compared to HCT116 cultured on plastic. Taken
together, these results suggest that HCT116 tumor cells are more heterogeneous on-chip,
recently corroborated by another tumor-on-a-chip model system (Hachey et al., 2021).
B.
E.
A.
Vimentin E-Cadherin HCT116 H2B GFP
Flow
Flow
Plated CRC Cells from Bottom Outflow
48 hr 120 hr
F.
C.
D.
HCT116 HT29
0.00
0.01
0.02
0.03
0.04
Ratio CRC Cells In Outflow /
Bottom Channel (D6)
****
Vimentin E-Cadherin
HCT116 H2B GFP
HCT116
2D Plastic
HCT116 CRC-Chip
Circulating Cells
0
2
4
6
8
10
Expression Relative to GAPDH
Vimentin E-Cadherin
✱✱✱✱
✱
HCT116
On-Chip Invaded
HCT116
2D Plastic
0
50
100
% Positive of Total Cells
E-Cadherin + Vimentin +
E-Cadherin + Vimentin + Neither
32
Figure 1.10. CRC cell heterogeneity during intravasation on-chip. (A) Representative images depict the
phenotypic heterogeneity of invaded HCT116 cells adhered to endothelial cells in the bottom chip channel. HUVECs
uniformly express vimentin (purple), while clusters of HCT116 cells (green) grow in tight aggregates and highly
express E-Cadherin (red) (top panels) or grow in more disperse colonies with higher vimentin (purple) expression
(bottom panels). Scale bar represents 200 μm and images are from day 6 of the experiment. (B) Representative
images depict the heterogeneity of HCT116 cells cultured on plastic. HCT116 cells (green) show moderate
expression of E-Cadherin (red), with relatively few vimentin-positive cells (purple) (white arrow). Cells were grown to
70% confluency before fixation, immunofluorescent staining, and imaging. Scale bar represents 200 μm. (C) HCT116
heterogeneity from (A) and (B) was quantitated using Perkin Elmer Harmony software. GFP+ tumor cells were
segmented and classified as E-Cadherin+ or vimentin+ based on intensity thresholds (N=5 Chips for chip
experiments, N=4 replicates for 2D plastic experiments). Data are represented as mean ± SEM. (D) Invaded CTCs
are found in the endothelial effluent where they are collected and cultured for down-stream analyses. (E) Viable
tumor cells were collected from the effluent of the bottom endothelial channel reservoir on day 6. Cells were plated
and counted via HCS imaging system once the cells had attached to the plate (6-10 hours later) (N=6 Chips for HT29
experiments, N=9 Chips for HCT116 experiments; ****p<0.0001). (F) RT-qPCR results show invaded HCT116 cells
have reduced E-Cadherin and increased vimentin expression compared to HCT116 cells grown on plastic tissue
culture dishes (N=6 replicates). Data are represented as mean ± SEM and analyzed using multiple unpaired t-tests;
****p<0.0001, *p<0.05).
Moreover, the microfluidic design of the chips allows for interrogation of any viable,
circulating cells that have perfused through the channel and accumulate in the outlet reservoir
(Figure 1.10D). I collected media from the outflow of the bottom channel every two days. HT29
cells did not appear in the bottom channel effluent, however I was able to isolate, quantify, and
relate the number of circulating HCT116 cells in the bottom channel outflow to the number of
invaded HCT116 cells still adhered to the endothelial channel (Figure 1.10E). These invaded,
circulating cells had lower E-cadherin and higher vimentin expression than HCT116s on plastic,
indicating these cells had acquired mesenchymal-like phenotypes (Figure 1.10F).
Initial Characterization of the Role of Mechanical and Biochemical Cues from the TME on
CRC Invasion
It is understood that cancer cells respond to mechanical cues present in the body (e.g.,
shear force from fluid flow) (Follain et al., 2020), however many of the mechanistic details
remain unknown as these processes are difficult to study. A major advantage of the CRC-Chip
model over other organs-on-chip technologies is the ability to mimic peristalsis, a physiological
process of muscle contraction and relaxation that naturally occurs in the colon. I showed that in
the presence of peristalsis (10% strain; 0.2 Hz), HCT116 tumor cell invasion increased
33
dramatically compared to static conditions (>3-fold) (Figure 1.11A). The increased invasive
phenotype in response to mechanical forces requires further study, including alternative
mechanisms beyond epithelial barrier disfunction, and will be extensively discussed in Chapter
3.
Figure 1.11. The TME influences tumor cell invasion. (A) CRC-Chips were cultured in the presence of cyclic
peristalsis-like mechanical strain and HUVECs (N=12 Chips), without cyclic strain and with HUVECs (N=12 Chips),
with cyclic strain and without HUVEC cells (N=12 Chips) or without cyclic strain or HUVECs (N=9 Chips). The
invasion ratio of the HCT116 cells was determined via microscopy. Data are represented as boxplots and analyzed
using a 1-way ANOVA with multiple comparisons; **p<0.01, ***p<0.001. (B) Representative CAFs derived from two
patients were labeled with Cell Tracker Deep Red and seeded in the top channel prior to epithelial cell and HCT116
cell seeding (N=4 replicates for each patient-derived CAF) and the invasion ratio was quantified on day 6. Data are
represented as boxplots and analyzed using a 1-way ANOVA with multiple comparisons; ***p<0.001; ****p<0.0001.
(C) Representative images of CAF000W8 and CAF000UE on day 0 (one day after tumor cell seeding) illustrate
heterocellular interactions on chip.
A.
C.
B.
HCT116 H2B GFP CAF000W8 CAF000UE
Control CAF000W8 CAF000UE
0
10
20
30
40
Fibroblast Co-culture
D6 Invasion Ratio
(Normalized to D0)
***
****
0
5
10
15
20
25
D6 Invasion Ratio
(Normalized to D0)
+
+
+
+
-
+
+
+
-
CRC
Peristalsis
Endothelial Cells
**
***
+
-
-
**
34
In addition to the physical forces present in the TME, heterocellular interactions can also
be examined using this model. I showed that cancer-endothelial cross talk is important for
driving an invasive phenotype. In the presence of HUVECs, the HCT116 invasion rate was
significantly higher than when HUVECs were not seeded in the CRC-Chip (Figure 1.11A). There
was no difference in HCT116 invasion ratio in the presence or absence of cyclic strain when the
HUVECs were absent, suggesting an important role of the endothelial cells in the invasive
phenotype in the presence of peristalsis (discussed further in Chapter 3). This data supports
published evidence that the tumor:blood vessel tissue:tissue interface is a critical modulator of
cancer progression (Amos and Choi, 2021; Choi and Moon, 2018; Nguyen et al., 2019).
Furthermore, CAFs, the most abundant cell type in the cancer stroma, have been
implicated in promoting invasion (Karnoub et al., 2007; Orimo et al., 2005) and cancer
metastasis (Sahai et al., 2020) and the physical interaction between CAFs and tumor cells has
been shown to be important for invasion (Gaggioli et al., 2007). In order to interrogate the role of
CAFs in CRC invasion in our system, I introduced a layer of CAFs to the CRC-Chip after
modification of the ECM in the top channel to ensure CAF attachment and viability (set-up ii of
Figure 1.4A, Table 1.1), leaving the endothelial channel unchanged. The CAFs were initially
seeded as a monolayer on the chip membrane coated with collagen IV but transitioned to form
extended networks after seeding the Caco2 C2BBe1 layer and the HCT116 tumor cells two
days later. The presence of the CAF000UE and CAF000W8 significantly increased invasion
(Figure 1.11B). Notably, confocal microscopy revealed that the physical presence of the CAFs
influenced how the HCT116 cells seeded onto the epithelium. The HCT116 cells clustered
around and on top of the CAF networks (Figure 1.11C). While the invasion rates were increased
approximately 2-fold in the presence of CAF CM or the physical CAFs (CAF000W8 and
CAF000UE), the HCT116 seeding morphology suggests the physical interaction between CAFs
35
and tumor cells is important and requires further study and will be discussed extensively in
Chapter 2.
Optimization of Liver Chips to Study CRC Extravasation
Greater than 50% of colon cancer tumors will metastasize to the liver (Valderrama-
Treviño et al., 2017). I optimized a liver chip developed by Emulate, Inc, comprised of
hepatocyte, endothelial, stellate, and Kupffer cells, with the goal of utilizing it to better
understand the metastatic capabilities of invaded HCT116 tumor cells from the CRC-Chip.
Primary human hepatocytes were seeded in the top, ECM-coated channel and overlayed with
Matrigel. A mixture of human non-parenchymal cells (NPC) consisting of liver sinusoidal
endothelial cells (LSECs), liver Kupffer cells, and stellate cells was cultured in the ECM-coated
bottom, vascular channel (Table 1.9; Figure 1.12A). I characterized the functionality of the liver
chip as previously described (Jang et al., 2019). The hepatocytes formed branched bile
canalicular networks lined by functional multidrug resistance-associated protein 2 (MRP2) efflux
transporters (Figure 1.12B). The LSECs displayed the multifunctional scavenger receptor
stabilin-1, which is expressed on liver endothelial cells (Figure 1.12B) (Shetty et al., 2018). In
addition, I measured albumin secretion by the hepatocyte cells over the course of 12 days in
culture. Albumin production peaked between days 6-9 at approximately 60 µg/day/10
6
cells
(Figure 1.12C), similar to previously published data (Jang et al., 2019).
Table 1.9. Chip locations specified for cell types and ECM composition used in the Liver-
Chip.
Channel Cell Type Cell Fluorescent
Tag
ECM
Bottom Endothelial LSEC Unlabeled Collagen I and fibronectin
Bottom Stroma
(Fibroblast)
Stellate Unlabeled Collagen I and fibronectin
Bottom Stroma (Immune) Kupffer Unlabeled Collagen I and fibronectin
Top Parenchymal Hepatocyte Unlabeled Collagen I and
fibronectin, Matrigel
overlay
36
Figure 1.12. Characterization of human Liver Chip. (A.) Human LSECs, Kupffer, and stellate cells were seeded
into an ECM-coated lower, non-parenchymal channel (left panel) and human hepatocytes were seeded into an ECM-
coated upper, parenchymal channel (right panel) and overlayed with Matrigel. (B.) LSECs stain positively for stabilin-
1 (left panel), while the hepatocytes express MRP-2 (right panel), indicating the functionality of the cell types on the
Liver Chips. (C.) Albumin production by the hepatocytes was measured from the parenchymal effluent and
quantitated by ELISA. Effluent was collected every three days over the course of the experiment (12 days).
In order to interrogate the ability of CRC tumor cells to colonize a healthy liver, I
designed an extravasation assay to monitor GFP-labeled CRC tumor cells flowing into the non-
parenchymal channel of the healthy human Liver Chip (Figure 1.13A). Briefly, liver chips were
prepared as described above. The day after the NPC were seeded and the liver chips were
connected to flow, HCT116 H2B-GFP tumor cells were introduced to the non-parenchymal
channel media inlet and allowed to flow overnight. The location of the HCT116 tumor cells were
monitored via confocal microscopy the day after HCT116 introduction and over the course of the
experiment (14 days) (Figure 1.13B). In order to increase the opportunity for extravasation,
several key steps were identified during the optimization stage of this assay. First, the HCT116
were resuspended in a 50% Percoll solution to encourage flow and negate cell settling within
the non-parenchymal channel during the overnight flowthrough. Second, the amount of FBS in
LSECs + Kupffer Cells
+ Stellate Cells
Hepatocytes
A.
B.
C.
D3 D6 D9 D12
0
20
40
60
80
Time (Days)
Albumin
(µg/day/10
6
cells)
Dapi Stabilin-1 Dapi MRP-2
37
the hepatocyte media was increased from 2% to 10% to ensure HCT116 survival in the
epithelial channel. Third, HCT116 tumor cells were seeded at different timepoints and it was
determined tumor cells should be introduced to the liver chips early in the experiment (D3;
Figure 1.13B) to ensure healthy hepatocytes.
HCT116 tumor cells were flown from the inlet reservoir into the bottom channel and
adhered to the NPC cells (Figure 1.14). Extravasation events were observed in which HCT116
invaded from the NPC compartment, through the ECM-coated porous membrane, and into the
parenchymal chamber to colonize the hepatocyte cells (Figure 1.14). Extravasation events were
rare, with many clusters of tumor cells adhering to the top layer of the NPC, but only a few cells
adhering and proliferating in the top channel (Figure 1.14). The application of this assay will be
further discussed in Chapter 3, where I use the on-chip extravasation assay to better
understand the role of mechanical, peristalsis-like forces on CRC progression.
38
Figure 1.13. CRC tumor cell extravasation on-chip assay. (A.) Schematic of tumor cell extravasation. CRC tumor
cells (green) grow as clusters in the top, epithelial channel of the CRC-Chip before invading into the bottom, vascular
channel. The invaded cells can be collected and suspended in the vascular channel inlet reservoir of the Liver Chips.
Invaded CRC tumor cells are then flown into the liver vascular channel, where the cells adhere to the NPCs and
extravasate into the hepatocyte cell layer. (B.) Timeline of tumor cell extravasation experiment. Normal liver chips are
prepared (Days 0-2) prior to an overnight HCT116 flowthrough (Days 3-4). Liver chips are imaged on Day 4 to identify
adhered CRC tumor cells and chips are maintained for the duration of the experiment, before end point imaging is
performed to identify extravasated HCT116 (here Day 14).
Image Chips
Day 2 Day 1 Day 0
Surface Activation
Apply ECM coating
Day 3
Hepatocyte Seeding
Overlay
Image Chips
Day 4
Stellate Seeding
LSEC Seeding
Kupffer cell Seeding
Initiate flow
HCT116 flowthrough
Day 14
Preparation & Maintenance Imaging Timepoints
Day 5-13
Refresh Inlet Media
A.
B.
39
Figure 1.14. CRC Tumor Cells Extravasate on a Healthy Human Liver Chip. HCT116 H2B-GFP were flown
through the bottom, non-parenchymal channel on day 3 (the day after NPC seeding and connection to flow). Chips
were fixed on day 14, stained with DAPI (blue) and imaged with confocal fluorescent microscopy. The bottom panels
show snapshots of the bottom channel at various Z-heights, indicating HCT116 H2B-GFP cells adhering to the top
(left panel) and bottom layers (right panel) of the NPC. A colony of extravasated HCT116 tumor cells are shown in the
top panels.
Hepatocyte DAPI HCT116 H2B GFP
HCT116 H2B GFP
Non-Parenchymal
Cells DAPI
150 μm
90 μm -55 μm
Day 14
Overlay
HCT116 H2B GFP
Non-Parenchymal
Cells DAPI
200 μm
200 μm
40
1.3 DISCUSSION
I expanded upon a previously characterized Intestine Chip (Kim et al., 2012) to generate a
CRC-Chip that encapsulates cancer-specific biomimetic microenvironments to advance our
knowledge in the CRC research domain. Some of the results presented here corroborate
findings from previous cancer progression studies as a metric to validate our model system.
However, I also demonstrate for the first-time a CRC-Chip that recapitulates many aspects of
the complex tumor microenvironment that are difficult to reconstruct in other model systems.
This has led to the discovery of novel preliminary findings, including the impact of peristalsis and
the epithelial:endothelial tissue:tissue interface on tumor cell intravasation. More mechanistic
studies on the role of peristalsis and tumor-stromal interactions will be discussed in future
chapters to explore the impact of these findings. Here I will discuss the initial studies that
demonstrate the utility of the microfluidics-based CRC-Chip for the study of CRC progression.
The current chip design can be used to mimic the progression of colon cancer in which a
polyp forms in the colonic crypts before eventually evolving into a cancerous lesion that grows
into the intestinal lumen (Dekker et al., 2019; Humphries and Wright, 2008). When I introduced
CRC tumor cells into the model, the presence of the epithelium-tumor boundary resulted in 3D
“hot spots” of tumor cells. There was also a measurable shift in metabolite profiles from the
collected effluent between the Intestine Chip and the CRC-Chip, which mapped to the colorectal
cancer disease state. Metabolic reprogramming has been extensively studied in cancer and is a
hallmark of cancer progression (Faubert et al., 2020). Specifically, my results are corroborated
by published metabolomics datasets that found alterations in the TCA cycle, urea cycle, and
amino acid metabolism in colorectal cancer patients when compared to healthy controls
(Farshidfar et al., 2016; Tan et al., 2013). Though additional studies are needed to address the
metabolic reprogramming that occurred in the CRC-Chip, the microfluidic attributes of the chip
41
technology will aid in identifying metabolic phenotypes that link to disease progression and
metabolic vulnerabilities that may be targeted in metastatic tumor cells.
In addition to effluent-based metabolomics studies confirming CRC pathway-specific
metabolites in the chip, it also revealed tumor cell intravasation as a feature of the model. This
finding was based on dynamic metabolite changes in the endothelial channel indicative of tumor
cell presence. This prompted me to further explore the capacity of our system to interrogate the
invasive potential of CRC tumor cells. An advantage of this model is the transparent nature of
the chip material, which supports dynamic confocal imaging of cell phenotypes. I was able to
quantitate substantial differences in invasion into the vascular channel between aggressive and
non-aggressive CRC tumor cells and identify a metabolic profile for an aggressive, invasive
CRC tumor. More work is needed to understand whether these metabolic changes are
influenced by tumor-endothelial cross talk or solely the presence of invaded tumor cells. In
addition, I was able to visualize cellular heterogeneity as tumor cells intravasated. However,
questions emerge as to why some invaded tumor cells remain as tightly adhered clusters to the
endothelial wall, whereas, others end up in circulation and presumably are the mediators of
distant metastases. Carefully interrogating the role of dynamic processes such as heterogeneity
in cancer progression has proved challenging due to the lack of appropriate models (Brabletz et
al., 2018), however this preliminary work suggests the CRC-Chip is a useful tool in elucidating
cellular plasticity, invasiveness, and cancer progression.
Furthermore, this work demonstrated I can tune mechanical and biochemical cues within
the TME to better understand early tumor cell dissemination. Previously, the role of mechanical
forces on cancer progression has been technically challenging to study in the laboratory. The
introduction of microfluidic-based systems has expanded our understanding of how fluid shear
forces influence tumor cell (particularly circulating tumor cells) survival in circulation and
extravasation at the metastatic site(s) (reviewed extensively in (Follain et al., 2020)). Previous
42
work studying metastasis in a zebrafish model suggests that tumor cells which form stable
adhesions with the endothelium are more likely to extravasate (Follain et al., 2018; Osmani et
al., 2019). The CRC-Chip is an advantageous model to better understand and potentially target
the metastatic capabilities of the tumor cells that adhere to the endothelial channel.
I next wanted to investigate whether the tumor cells in the endothelial channel have a
propensity to extravasate and seed distant organs, such as the liver, given its frequency as a
metastatic site for CRC. Emulate, Inc. has developed a healthy human liver-chip comprised of
endothelial, stromal, immune, and parenchymal cell types, which I adapted to study how CRC
cells extravasate in a healthy liver setting. HCT116 tumor cells successfully extravasated from
the non-parenchymal channel into the hepatocyte cell layer. Other researchers have designed
microfluidic devices to study these later steps in the metastatic cascade. For example, one
group connected a colon cancer chamber to organ-specific epithelial cell chambers (i.e. lung or
liver) to study metastatic organotropism (Aleman and Skardal, 2019; Skardal et al., 2016). The
application of this extravasation assay will be discussed in later chapters.
The Organ Chip field is expanding in popularity and demonstrated utility; however, it is
still in its infancy especially in the context of cancer. Conversely, organoids have revolutionized
the cancer field because they are thought to better replicate organ complexity and function and
have been shown to predict patient anticancer drug response in early in vitro drug screens
(Driehuis et al., 2019; Drost and Clevers, 2018; Ooft et al., 2019; Vlachogiannis et al., 2018;
Yao et al., 2019). Nevertheless, the organoid model lacks aspects of the TME discussed here,
such as stromal-tumor interactions, tumor:endothelial tissue:tissue interfaces, and physical
forces. I demonstrated the utility of combining organoids and the Organ Chip technology to
study tumor cell progression as evidenced by patient-derived tumor organoids invading into the
endothelial channel using our CRC-Chip.
43
In conclusion, the CRC-Chip provides a human-relevant model system to examine CRC
progression within the TME milieu. The ability to monitor key steps in cancer metastasis is
unparalleled to other preclinical cancer models. This system will be critical to better understand
the mechanisms surrounding CRC early metastatic spread and potentially elucidate novel
therapeutic targets within the TME.
Limitations of the Study
I use a combination of primary, patient-derived cells and immortalized cell lines to build the
CRC-Chip. In this case, I believe the increased complexity of the microfluidic organ-on-chip
system more closely resembles in vivo tissue structure and function and will greatly advance
CRC mechanistic studies, however I acknowledge that the use of cell lines is a limitation of this
study, especially the use of the C2BBe1 clone of the Caco2 cell line. Although this cell line has
been selected because it is more representative of the human colon, I am working to include
patient-derived normal colon epithelial cells to this model system. Patient-derived cell
populations require further optimizations and, as such, the patient-derived organoid model
described here is exclusively cancer organoids. Future work will include additional primary cell
types from diverse patient populations to better represent the heterogeneity of human cancer
biology. In addition, since this Organ Chip platform is not currently compatible with live, dynamic
imaging, the use of static time-points in this study is a limitation. As the technology improves, I
will be able to monitor cell behavior in real time on-chip to better understand cancer
progression. Furthermore, while I suggest our CRC-Chip is capable of establishing an intact
barrier based on ZO-1 expression and permeability assays, an important caveat to this finding is
the lack of transepithelial/transendothelial electrical resistance (TEER) measurements of the
tight junction integrity.
44
1.4 FUTURE DIRECTIONS
Most of the work here describes a CRC-Chip with immortalized cell lines (discussed
above in the limitations of the study section). Current and future experiments are focused on the
utilization of patient-derived organoids on-chip. Expanding the work presented here to
encompass organoids from various CRC patients will give us greater insights into inter-tumor
heterogeneity. In addition, the tissue-tissue interface of the Organ Chips also supports the
performance of drug treatment studies in a physiological manner by flowing drugs through the
endothelial channel and measuring tumor cell response in the epithelial channel as seen in
previous reports (Carvalho et al., 2019). Further development of the patient-derived CRC-Chip
to include autologous normal and tumor organoids will pave the way for studying cancer
progression in a patient-specific manner and may enhance precision medicine approaches
(Ramzy et al., 2020).
1.5 METHODS
Cell Culture
Commercially Available Cell Lines: Human umbilical vein endothelial cells (HUVEC) expressing
Red Fluorescent Protein (RFP) (Angio-Proteomie, #cAP-0001RFP) were expanded in EBM-2
media with EGM-2 SingleQuots Supplements (2% FBS, 1% Pen-Strep, Hydrocortisone, hFGF-
B, VEGF, R3-IGF-1, Ascorbic Acid, hEGF, and heparin in proprietary concentrations) (Lonza
#CC-3162; supplemented with 1% Penicillin-Streptomycin (Pen-Strep) in lieu of Gentamicin).
Caco2 C2BBe1 cells (ATCC, #CRL-2102) were grown in DMEM (Gibco, #10569-010) with 10%
fetal bovine serum (FBS) and 1% Pen-Strep. HCT116 and HT29 cells (ATCC #CCL-247 and
#HTB-38) were grown in McCoy’s 5A media (Gibco, #16600-082) with 10% FBS and 1% Pen-
Strep, labeled with LentiBrite Histone-H2B-GFP Lentiviral Biosensor (Millipore, #17-10229), and
sorted to achieve a pure fluorescent population. CCD18Co (ATCC, #CRL-1459) cells were
45
grown in Eagle’s Minimum Essential Medium (Corning, #10-009-CV) with 10% FBS and 1%
Pen-Strep. All cells were cultured under standard laboratory conditions (5% CO2, 37
o
C).
Patient-Derived Samples: Tissue resections were received from the USC Norris Comprehensive
Cancer Center following Institutional Review Board (IRB) approval (Protocol HS-06-00678;
approval date 08-02-2019) and patient consent. Tumor profiles, including known tumor
mutations, sex, and treatment information, are detailed in Table S10. Human primary fibroblasts
and organoids were derived from CRC tumors via a previously described method (Sato et al.,
2011; Sato et al., 2009). Briefly, tumor pieces were minced and enzymatically digested using
1.5 mg mL
-1
collagenase (Millipore, #234155), 10 µM LY27632 (Millipore, #5.09228.0001) and
20 µg mL
-1
hyaluronidase (MP Biomedicals, #0210074080) for 30 minutes at 37
o
C. The
resulting cell mixture was either plated in basement membrane extract (BME; Cultrex, #3533-
005-02) to derive organoids or on plastic tissue culture plates to select for fibroblasts. Organoids
and fibroblasts were cultured in a defined colon media (ADMEM/F12 (Gibco, #12634-010), 10%
FBS, 1% Pen-Strep, supplemented with 100 ng mL
-1
Noggin (Tonbo, #21-7075-U500), 50 ng
mL
-1
epidermal growth factor (EGF) (Life Technologies, #PHG0313), 10 μM SB202190 (Sigma
Aldrich, #S7067), 500 nM TGF-β RI Kinase Inhibitor IV (A83-01) (Millipore, #616454-2MG), 10
mM Nicotinamide (Sigma Aldrich, #N0636), 1x B27 (Gibco, #17504-001), 1mM N-acetylcysteine
(Sigma Aldrich, #A9165), 1x N2 (Gibco, #17502-048), 1x HEPES (Gibco, #15630-080), 1x
GlutaMax (Gibco, #35050-061)). After establishment and expansion of organoids, they were
subsequently labeled with H2B-GFP lentivirus. Cells cultured on plastic tissue culture plates
were confirmed to be cancer-associated fibroblasts (CAFs) and were used for experiments
between passage 3 and passage 7.
46
RT-qPCR
To interrogate epithelial or mesenchymal marker expression in invaded tumor cells from the
CRC-Chips, I performed RT-qPCR analysis on cultured HCT116 collected from the chip
effluent. Cellular RNA was extracted using RNAspin Mini RNA Isolation Kit (GE Healthcare;
#25-0500-71) and cDNA was reverse transcribed using iScript Reverse Transcription Supermix
(Bio-Rad, #1708841) following manufacturer’s instructions. The cDNA was then amplified using
iScript SYBR Green Master (Bio-Rad; #1708880). The sequences for PCR primers are listed in
Table 1.10. Results were normalized to GAPDH expression for all experiments.
Table 1.10. Gene specific primers for qPCR. Related to Figure 1.10.
Gene Direction Sequence 5’ à 3’
Human GAPDH Forward TCTGGTAAAGTGGATATTGTTG
Reverse GATGGTGATGGGATTTCC
Human E-Cadherin Forward TTTGTACAGATGGGGTCTTGC
Reverse CAAGCCCACTTTTCATAGTTCC
Human Vimentin Forward GAGAACTTTGCCGTTGAAGC
Reverse GCTTCCTGTAGGTGGCAATC
Tumor Cell Transwell Invasion Assay
Fluoroblok
TM
transwell inserts with PET membranes (6.5 mm membrane diameter, 8 μm
membrane pore size, 0.3 cm
2
cell culture area, Corning, #351152) were coated with 30 μg mL
-1
type I collagen (Corning, #C354249) and 100 μg mL
-1
Matrigel (Corning, #356231) for 2 hours at
37
o
C before the ECM was gently aspirated from the insert. For transwell experiments with
endothelial cells, the inserts were inverted and seeded with HUVECs (2.1x10
5
cells in 35 μL;
7x10
5
cells cm
-2
) and incubated at 37
o
C for 2 hours. The inserts were flipped over, placed in
wells with 500 μL endothelial media in the bottom chamber, and Caco2 C2BBe1 cells were then
47
seeded on top (1.1x10
5
cells in 200 μL; 3.7x10
5
cells cm
-2
). CRC tumor cell lines (HCT116 H2B-
GFP or HT29 H2B-GFP) were added (3.6x10
4
cells in 100μL; 1.2x10
5
cells cm
-2
) 48 hours later.
Cells were incubated in the same cell culture media as organ-chip experiments in order to
facilitate comparisons between experiments. Caco2 C2BBe1 and CRC tumor cells in the top
chamber were maintained in DMEM with 10% FBS and 1% Pen-Strep. Fully supplemented
EBM-2 media (2% FBS, 1% Pen-Strep, Hydrocortisone, hFGF-B, VEGF, R3-IGF-1, Ascorbic
Acid, hEGF, and heparin in proprietary concentrations) was placed in the bottom of the wells.
The transwells were imaged the day after tumor cell seeding (D0) and 48 hours later (D2) using
the Perkin Elmer Operetta High Content Screening (HCS) platform. The number of GFP
+
cells in
the bottom chamber were quantified using Perkin Elmer Harmony software.
Tumor Cell Immunofluorescence in 2D
HCT116 H2B-GFP or HT29 H2B-GFP cells were seeded on 24-well cell culture plates and
allowed to grow to 70% confluency. Cells were washed with PBS and fixed with 4%
paraformaldehyde (Electron Microscopy Sciences, #15710), incubated for 15 minutes, and
permeabilized with 1% saponin (Sigma Aldrich, #84510). Blocking buffer of 2% bovine serum
albumin (BSA) (Millipore, #260-500GM) and primary antibodies were incubated overnight at 4
o
C
before a 2-hour incubation with secondary antibodies (1:100, Molecular Probes, Invitrogen,
#A21428 and #A21235) diluted in blocking buffer. The primary antibodies used for these studies
were anti-E-cadherin (1:25; Abcam, #ab1416) and anti-vimentin (1:50; Abcam, #ab92547). Cells
were imaged using the Perkin Elmer Operetta CLS High Content System. Quantitation of E-
Cadherin+ and Vimentin+ tumor cells (identified by GFP) was performed using the Perkin Elmer
Harmony software.
Microfluidic Organ-Chip Design and Culture
48
Chips were acquired from Emulate, Inc. The fabrication methods have been previously
described (Huh et al., 2013). In brief, the chips are made of transparent elastomeric polymer
(polydimethylsiloxane, PDMS). They are divided into upper (1 mm high x 1 mm wide) and lower
(0.2 mm high x 1 mm wide) microfluidic compartments separated by a thin porous membrane
(50 µm thick with 7 µm diameter pores; 17.1 mm
2
co-culture region). The upper compartment
hosts epithelial cells (plus additional cell types), and the lower compartment hosts the
endothelial cells. Each compartment is coated with a tissue-specific ECM prior to cell seeding.
The chips are attached to a Pod
TM
portable module that encloses the chips to control sterility,
holds inlet cell culture media and effluent, allows for monitoring via microscopy, and is designed
to ensure no pressure differentials between channels. The chips and pod are then housed in an
automated culture module instrument (Zoë
TM
culture module and Orb
TM
hub module, Emulate,
Inc.) that controls the fluid flow and stretching forces while inside an incubator.
Caco2 Intestine Chip: Methods for establishing the Intestine Chip cell culture have been
previously described (Kim et al., 2012). Briefly, the chip PDMS membranes were activated by
proprietary Emulate Reagents 1 and 2 (Emulate, Inc, ER-1 and ER-2) under UV light for 20
minutes. The epithelial and endothelial channels were coated with a mixture of 30 μg mL
-1
type I
collagen (Corning, #354249) and 100 μg mL
-1
Matrigel (Corning, #356231) for 2 hours at 37
o
C
before washing with PBS. RFP-labeled HUVEC cells were seeded into the bottom channel
(1.2x10
5
cells in 20 μL; 7x10
5
cells cm
-2
). The chips were inverted and incubated at 37
o
C for 2
hours. The density of HUVEC cells was optimized to ensure the cells would self-assemble a
tube along the channel. After HUVEC attachment, Caco2 C2BBe1 cells were seeded into the
top channel (62,500 cells in 50 μL; 3.7x10
5
cells cm
-2
) and the chips were incubated overnight at
37
o
C. The chips were perfused with DMEM, 10% FBS, 1% Pen-Strep in the top channel and
endothelial media (EBM-2 fully supplemented with 2% FBS, 1% Pen-Strep, Hydrocortisone,
hFGF-B, VEGF, R3-IGF-1, Ascorbic Acid, hEGF, and heparin in proprietary concentrations) in
49
the bottom channel at 30 μL hour
-1
(0.02 dyne cm
-2
) starting the day after cell seeding. Cyclic,
peristalsis-like membrane deformations (10% strain, 0.2 Hz) were also initiated the day after cell
seeding using an electronic vacuum pump system (Emulate, Inc).
Caco2 + CRC Cell Lines: To modify the Intestine Chip to model CRC, I introduced CRC cells on
top of the Caco2 C2BBe1 epithelial cell layer. Forty-eight hours after Caco2 cell seeding, CRC
cells (HCT116 H2B-GFP or HT29 H2B-GFP) were seeded in the top channel (2x10
4
cells in 50
μL; 1.2x10
5
cells cm
-2
). The chips were incubated for 2 hours at 37
o
C after which 30 μL hour
-1
(0.02 dyne cm
-2
) flow was re-established with DMEM, 10% FBS, 1% Pen-Strep in the top
channel and endothelial media (EBM-2 fully supplemented with 2% FBS, 1% Pen-Strep,
Hydrocortisone, hFGF-B, VEGF, R3-IGF-1, Ascorbic Acid, hEGF, and heparin in proprietary
concentrations) in the bottom channel. Cyclic, peristalsis-like membrane deformations (10%
strain, 0.2 Hz) were initiated the day after HCT116 seeding.
Caco2 + CRC Cell Lines + Patient-Derived Fibroblasts: Fibroblasts were introduced to the
epithelial channel based on methods developed by Emulate, Inc. Briefly, the top channel was
coated with 200 μg mL
-1
type IV collagen (Sigma Aldrich, #C5533) and the bottom channel was
coated with 100 μg mL
-1
Matrigel (Corning, #356231) and 30 μg mL
-1
type I collagen (Corning,
#354249). HUVECs were seeded on the bottom channel as described above. Fibroblasts were
labeled with Cell Tracker Deep Red (1:1000; Invitrogen, #C34565) and seeded onto the top
channel (7.5x10
4
cells in 50 μL), incubated for at least 4 hours, and then overlaid with a 100 μg
mL
-1
Matrigel (Corning, #356231) and 30 μg mL
-1
type I collagen (Corning, #354249) layer and
incubated for an additional 2 hours. The Caco2 C2BBe1 and HCT116 cells were then seeded
as described above. Chips were perfused with 1:1 Colon media: DMEM with 10% FBS and 1%
Pen-Strep in the top channel and endothelial media (EBM-2 fully supplemented with 2% FBS,
1% Pen-Strep, Hydrocortisone, hFGF-B, VEGF, R3-IGF-1, Ascorbic Acid, hEGF, and heparin in
proprietary concentrations) in the bottom channel.
50
Patient-Derived Colon Cancer Organoids: The above CRC-Chip protocol was amended by
introducing human colon tumor-derived organoids to the top channel. Briefly, the top channel
was coated with 250 μg mL
-1
Matrigel (Corning, #356231) and the bottom channel was coated
with 100 μg mL
-1
Matrigel (Corning, #356231) and 30 μg mL
-1
type I collagen (Corning,
#354249). RFP-labeled HUVECs were seeded on the bottom channel as described above.
Organoids were incubated in Gentle Cell Dissociation Reagent (STEMCELL Technologies,
#07174) on ice before enzymatically dissociated using 50% TrypLE (Gibco; 12605-028)
supplemented with 10 mM LY-27632 (Millipore, #5.09228.0001) and seeded on the top channel
as described previously (Kasendra et al., 2020). Chips were incubated overnight before starting
flow and mechanical deformations as noted above. Chips were perfused with fully
supplemented Colon media in the top channel and endothelial media (EBM-2 fully
supplemented with 2% FBS, 1% Pen-Strep, Hydrocortisone, hFGF-B, VEGF, R3-IGF-1,
Ascorbic Acid, hEGF, and heparin in proprietary concentrations) in the bottom channel.
Liver Chip: Methods for establishing the normal human liver-chip have been previously
described (Jang et al., 2019) and are available from Emulate, Inc. Briefly, chips were activated
as described above and the channels were coated with collagen type 1 (100 µg/mL; Corning,
#354249) and bovine fibronectin (25 µg/mL; Gibco, #22010-018). Primary hepatocytes (source)
were seeded in the top channel at a concentration of 3.5 million cells/ml and were overlayed
with Matrigel (250 µg/mL; Corning, #354234) 24 hours after hepatocyte seeding. After the
overlay incubated overnight, the bottom channel could be seeded with endothelial cells or a
mixture of non-parenchymal cells (human liver sinusoidal endothelial cells (LSECs), Kupffer
cells, and stellate cells). For co-culture experiments (hepatocytes and LSECs), LSECs were
seeded in the bottom channel at 3-4 million cells/ml. For quad-culture experiments
(hepatocytes, LSECs, Kupffer, and stellate cells), non-parenchymal cells were seeded at the
51
following concentrations: LSECs at 3 million cells/ml, Kupffer cells at 0.5 million cells/ml, and
stellate cells at 0.1 million cells/ml. The chips were perfused (30 µL/hour) with the following
media: top channel - William’s E medium (Sigma, #W4128) supplemented with GlutaMAX
(Gibco, #35050-061), ITS+ Premix containing human recombinant insulin, human transferrin,
selenous acid, bovine serum albumin, and linoleic acid (Corning, #354352) dexamethasone
(Sigma-Aldrich, #D4902), ascorbic acid (Sigma-Aldrich, #5960), fetal bovine serum, and
penicillin/streptomycin; bottom channel – CSC endothelial medium (Cell Systems, #4Z3-500)
supplemented with culture-boost supplement packs (Cell Systems, #4CB-500), Pen/strep, and
FBS.
On-Chip Permeability Assay
The barrier function of the intestinal epithelium was measured by adding Inulin-Fluorescein
isothiocyanate (FITC, 2-5kDa; Sigma Aldrich, #F3272) to the epithelial channel at the start of
fluid flow and maintained for the duration of the experiment. The effluent from the bottom
channel was collected every two days and the fluorescence was measured using a plate reader
(Spectramax). Apparent permeability (Papp) was calculated using the following formula:
𝑷
𝒂𝒑𝒑
=
𝑪
𝒐𝒖𝒕𝒑𝒖𝒕
× 𝑭𝒍𝒐𝒘 𝑹𝒂𝒕𝒆
𝑪
𝒊𝒏𝒑𝒖𝒕
× 𝑨
where Coutput is the concentration of Inulin-FITC in the endothelial outflow, A is the seeded area,
and Cinput is the concentration of Inulin-FITC flowed into the epithelial channel. Duplicate or
triplicate chips per condition were performed for a set of three separate experiments.
52
CRC-Chip and Liver Chip Immunofluorescence
CRC Organ Chips were manually washed by flowing PBS through the endothelial and epithelial
channels. The chips were fixed with 4% paraformaldehyde (Electron Microscopy Sciences,
#15710), incubated for 15 minutes, and permeabilized with 1% saponin. Blocking buffer of 2%
bovine serum albumin (BSA) and primary antibodies were incubated overnight at 4
o
C before a 2
hour incubation with secondary antibodies (1:100, Molecular Probes, Invitrogen, #A21428 and
#A21235) diluted in blocking buffer. The primary antibodies used for the CRC-Chip studies were
anti-E-cadherin (1:25; Abcam, #ab1416), anti-VE-cadherin (1:25; Abcam, #ab33168), anti-
vimentin (1:50; Abcam, #ab93547), anti-ZO-1 (1:100; Thermo-Fisher, #339194). The primary
antibodies used for the Liver Chip studies were anti-MRP-2 (1:100, Abcam, #ab3373) and anti-
stabilin (Novus Biologicals, #NB1-84444). DAPI (Sigma-Aldrich, D9542) was used to label all
nuclei. Chips were imaged using the Perkin Elmer Operetta CLS High Content System.
Quantitation of E-Cadherin+ and Vimentin+ (tumor cells identified by GFP) was performed using
the Perkin Elmer Harmony software based on intensity thresholds.
Tumor Cell Invasion Assay On-Chip
CRC-Chips were prepared as described above. Tumor cell behavior was monitored via live cell
imaging using a confocal laser-scanning microscope (Olympus FV3000). Imaging began the
day after CRC cell seeding (day 0) and the same regions were imaged on subsequent imaging
days. Images (10x magnification) were acquired at six separate regions across the epithelial
channel (in the xy-direction), starting from the bottom of the endothelial channel up to the top of
the epithelial channel (roughly 400 μm) at 5 μm increments (in the z-direction). Image z-stacks
were imported into Imaris image analysis software for quantification of invasion. Images were
manually divided into top (above the membrane) and bottom (below the membrane) channel
regions. The Imaris cell detection algorithm was used to detect the number of GFP
+
cells in
53
each region. An invasion ratio was calculated based on the total number of GFP
+
cells in the
bottom channel divided by the total number of GFP
+
cells in the top channel at specified time
points. Raw invasion ratios were normalized to D0 to account for any outliers showing more
invasion than typical on D0 (shown in Figure 1.8A).
Inlet (DMEM with 10% FBS and 1% Pen-Strep for the top channel and fully supplemented EBM-
2 media (2% FBS, 1% Pen-Strep, Hydrocortisone, hFGF-B, VEGF, R3-IGF-1, Ascorbic Acid,
hEGF, and heparin in proprietary concentrations) for the bottom channel) and effluent media
was collected from the media reservoirs at the time of imaging and spun down at 900 rpm (200
rcf) for 5 minutes. The supernatant was collected for mass spectrometry-based metabolomic
analysis (see below) and the pelleted cells were resuspended and plated into a 96-well plate.
After cell attachment, the cells were stained with 5 μg mL
-1
Hoechst 33342 (nuclear dye;
Invitrogen, #H1399) and the cells were imaged via the Operetta HCS. The numbers of GFP
+
nuclei were quantified via the Harmony software.
Mass Spectrometry-Based Metabolomics of CRC-Chip
Metabolite Extraction: 100 μL of each sample was extracted with 500 μL extraction solvent
(80:20 Methanol:Water) spiked with 10 μL internal standard mix (Cambridge Isotopes
Laboratories Inc. Metabolomics QC Kit, #MSK-QC2), vortexed briefly and sonicated for 1 min to
precipitate the proteins. After storing for 60 min at −20 °C, the samples were centrifuged at
13,000 x g for 30 min at 4 °C. 450 μL of supernatant was transferred into a microcentrifuge tube
and dried using vacuum centrifugation at room temperature (approximately 2 hours). The dried
samples were resuspended in 100 μL resuspension solvent (50:50:0.1 (v/v)
Water:Methanol:Formic Acid) with 10 μL internal standard (Cambridge Isotopes Inc.
Metabolomics QC Kit, #MSK-QC1) spiked in and vortexed briefly to mix.
54
Liquid Chromatography with Tandem Mass Spectrometry (LC-MS/MS): Extracted metabolite
were analyzed using an ultra high performance liquid chromatography (UHPLC) system 1290
connected to a quadrupole time of flight (Q-TOF 6545) mass spectrometer from Agilent
Technologies (Santa Clara, CA, USA) equipped with an orthogonal DUAL AJS-ESI interface.
Samples were subjected to both hydrophilic interaction liquid chromatography (HILIC)
separation (Agilent Poroshell-HILIC-p 2.1 X 100, 2.7μm; Agilent Technologies, #675775-924)
and reverse phase C18 separation (Agilent Zorbax RRHD eclipse plus C18 column 2.1 X 50,
1.8μm; Agilent Technologies, #959757-902) and data were either collected in positive or
negative ion mode. Data were acquired from 50 to 1250 m z
-1
at 2 spectra s
-1
. Electrospray
ionization (ESI) source conditions were set as follows: gas temperature 290 °C, drying gas 9 L
min
-1
, nebulizer 35 psi, fragmentor 125 V, sheath gas temperature 350 °C, sheath gas flow 11 L
min
-1
, nozzle voltage 1000V.
For HILIC chromatographic separation, a two-solvent gradient flowing at 0.5 mL min
-1
(Mobile
Phase A: 10mM Ammonium Acetate in 90:10 Water:Acetonitrile, pH 9.2, 5 μM Agilent
InfinityLab Deactivator Additive (Agilent, #5191-4506) per vendor instruction, Mobile Phase B:
10mM Ammonium Acetate in 90:10 Acetonitrile:Water, pH 9.2, 5 μM Agilent InfinityLab
Deactivator Additive per vendor instruction) was used. Column was equilibrated at 98% B for 3
minutes and a sample was introduced. The solvent ratio was then reduced from 95% B to 50%
B over 10 minutes and then brought back up to 98% B over 5 minutes. For reverse phase C18
chromatographic separation, a two-solvent gradient running at 0.3 mL min
-1
(Mobile Phase: A:
100:0.1 Water:Formic Acid, B: 100:0.1 Methanol:Formic Acid) was used. Column was
equilibrated at 2% B for 1 minute and a sample was introduced. The solvent ratio was then
increased from 2% B to 98% B over 15 minutes and then reduced back to 2% B over 4
minutes. Both HILIC and reverse phase C18 had injection volumes of 2 μL and column
temperature of 30 °C.
55
Metabolomics Data analysis: The LC-MS/MS data acquired using Agilent Mass Hunter
Workstation (“.d” files) were processed in Profinder (Agilent Technologies) for batch recursive
extraction of features. Spectral peak extraction was performed with a minimum peak height of
500 counts and charge state of one. Further, retention time and mass alignment corrections
were performed on the runs to remove non-reproducible signals. The resulting features were
then exported as “.cef” files to Mass Profiler Professional (MPP) software (Agilent Technologies,
Santa Clara, CA, USA) for multivariate analysis. Principal Component Analysis (PCA) was
performed to check the quality of the samples and then the data containing filtered features
were processed by unpaired t-test to find the difference between the groups. Only analytes with
p values < 0.05 and fold change > 2 were regarded as statistically significant. Additionally,
multiple test correction using Bonferroni was applied to reduce false positives and false
negatives in the data. The statistically filtered data was then exported to Mass Hunter for
targeted MS/MS analysis. All target metabolites were identified by matching their m/z values
and retention times with in-house libraries (PCDL, Agilent technologies) of measured values for
the IROA technologies MSMLS compound library. Metabolites that did not match to our internal
library were matched against METLIN library or were listed with a predicted molecular formula.
Pathway Analysis: To visualize the metabolites within relevant networks of pathways, the
detected statistically significant metabolites were mapped to the curated pathways using
Ingenuity Pathway Analysis (IPA) software (digitalinsights.qiagen.com).
Liver Chip Albumin ELISA
Human albumin secretion from the upper channel was quantified using ELISA kits (Abcam,
#ab179887) to test for the functionality of hepatocyte cells. Effluent media from the top channel
was collected over the course of the experiment and the assays were performed according to
the manufacturer’s instructions.
56
Liver Chip Extravasation Assay
HCT116 tumor cells were flown through the bottom channel of the liver in order to mimic
circulating tumor cells seeding at a metastatic site. Briefly, liver chips were seeded as described
above. In order to promote extravasation, the top channel media was perfused with hepatocyte
media as described above with an increased amount of FBS (10%). HCT116 tumor cells were
introduced to the bottom channel of the liver chips the day after connection to flow. Briefly,
HCT116 cells (10,000 cells/chip) were resuspended in Percoll Solution (Sigma-Aldrich,
#P4917):NPC media at a 1:1 ratio and flown through the bottom channel at a rate of 600 µL/hr
for 10 minutes. After the flush cycle, the flow rate was reduced to a rate of 60 µL/hr for
overnight. NPC media was replenished the following morning and the flow rate was returned to
30 µL/hr.
Tumor cell behavior was monitored via live cell imaging using a confocal laser-scanning
microscope (Olympus FV3000). Imaging began the day after CRC cell seeding (day 0) and the
same regions were imaged on subsequent imaging days. Images (10x magnification) were
acquired at six separate regions across the channel (in the xy-direction), starting from the
bottom of the endothelial channel up to the top of the upper channel (roughly 400 μm) at 5 μm
increments (in the z-direction). Image z-stacks were imported into Imaris image analysis
software for quantification of invasion. Images were manually divided into top (above the
membrane) and bottom (below the membrane) channel regions. The Imaris cell detection
algorithm was used to detect the number of GFP
+
cells in each region and any GFP+ cells in the
top channel marked any extravasation events in that particular chip.
57
Statistical Analysis
Unless otherwise noted, all experiments were performed with 2-3 chips per condition and
repeated independently at least 3 times. Analysis of variance (ANOVA) and t-tests were
performed using GraphPad Prism 8 software, with the p-value < 0.0001: ****; p-value < 0.001:
***; p-value < 0.01: **; p-value < 0.05: *, as noted in all figure legends. Data is reported as mean
with standard error of the mean (SEM) or as boxplots when noted, with boxplots designating the
25
th
-75
th
percentiles, median value, minimum value, and maximum value.
58
CHAPTER 2
Colorectal CAFs influence early metastatic events in patient-
specific manner
Associated publications:
Carly Strelez, Sujatha Chilakala, Kimya Ghaffarian, Roy Lau, Erin Spiller, Nolan Ung, Danielle
Hixon, Ah Young Yoon, Ren X. Sun, Heinz-Josef Lenz, Jonathan E. Katz, and Shannon M.
Mumenthaler. Human colorectal cancer-on-chip model to study the microenvironmental influence
on early metastatic spread. iScience 2021.
Carly Strelez, Curran Shah, Kimya Ghaffarian, Shannon M. Mumenthaler. Diversity of cancer-
associated fibroblasts and their influence on cancer progression. In Press at Physical Biology
2021.
Contributions:
CAF cultures were maintained by C. Strelez and K. Ghaffarian. qPCRs were performed by C.
Strelez and K. Ghaffarian. CAF secretome analyses were performed by C. Strelez, R. Sun, and
N. Ung. All on-chip invasion assays were performed and analyzed by C. Strelez. RNA-seq
analysis was performed by N. Paidimukkala.
2.1 INTRODUCTION
Cancer-associated fibroblasts (CAFs), the dominant stromal cell type within the tumor
microenvironment (TME), have been linked to several tumor promoting mechanisms across
cancer types, including increased tumor cell proliferation and invasion, and protection against
drug-induced apoptosis (Sahai et al., 2020). Research surrounding CAFs has historically
focused on identifying markers that uniquely define this cell population. Several studies have
attempted to target CAF-specific markers in in vivo models and clinical trials, yet these attempts
have been unsuccessful (Sahai et al., 2020). Of late, CAF studies have evolved to include
sophisticated subpopulation analyses, revealing significant intra- and inter-tumoral CAF
heterogeneity. The presence of CAF diversity may explain why the identification of CAF-specific
targets has been challenging and unsuccessful in clinical trials. Increasingly, research in the
field suggests that certain CAF populations are indicative of poor patient prognosis, confer drug
59
resistance, and increase tumor invasion and metastasis, while other CAF populations are
capable of restraining tumor growth, stimulating a pro-inflammatory TME, and predicting
response to immunotherapy (Biffi and Tuveson, 2021). While CAF heterogeneity is
acknowledged, there remains limited insight into the functional implications of this heterogeneity
on cancer progression.
Important work is being done to identify CAF subpopulations, yet there is a critical need
to connect these findings to a direct impact on tumor cell behavior. To tackle this challenge, we
must address fundamental questions such as: What cues from genetically diverse cancer cells
or the TME are responsible for CAFs switching cell states? How do CAF phenotypic states
influence cancer progression? To respond to these questions, it is imperative to recreate and
tune aspects of the TME and measure the resulting changes to CAFs and cancer cells (Figure
2.1). This requires (1) biological model systems in which, at a bare minimum, cancer cells and
CAFs can be physically cultured together, (2) the ability to deconvolve CAF and cancer cell
behavior, (3) measurements that capture spatial and temporal dynamics, and (4) validation of
findings with human tumor tissue.
In this chapter, I will discuss the heterogeneity observed in patient-derived CAFs and
delve further into how I can use the CRC-Chip platform outlined in Chapter 1 to better
understand how this heterogeneity drives CRC tumor progression.
60
Figure 2.1. CAFs are comprised of multiple subpopulations that can interconvert based on the cues from the TME.
CAF subtypes differentially influence various aspects of tumor cell behavior.
2.2 RESULTS
CRC CAF Characterization
CAFs were isolated from patient tumors and cultured in vitro. CAFs used throughout this
dissertation were derived from patient tumors spanning the CRC staging classification, as
shown in Table 2.1. Cells were determined to be CAFs based on morphology (Figure 2.2A) and
the expression of known CAF markers (aSMA, vimentin, and fibronectin) (Figure 2.2B). Initial
characterization of patient-derived CAFs showed that the CAFs had variable morphology and
expression of the known CAF markers aSMA, vimentin, and fibronectin (Figure 2.2).
61
Table 2.1. Clinical details of patient samples.
Cell
Model
Sample
ID
Sex CRC
Stage
Resection
Site
Molecular
Information of
Patient Tumor
Clinical Treatment
CAF 000U8 F I Colon Microsatellite
stable (MSS)
Unknown
CAF 000UM F I Colon MSS, BLM,
DDB2
No treatment
CAF 000UH M I Colon Not available No treatment
CAF 000UU F I Colon Not available No treatment
CAF 000US M IIB Colon KDR, KRAS,
TP53, SMAD4
mutant
Adjuvant Xeloda
CAF 000U0 M IIC Colon PTEN, PIK3CA,
CTNNB1
mutant
No treatment
CAF 000UE F IIIB Colon MSS Adjuvant FOLFOX
CAF 000VT M IIIB Colon MSS, KRAS
mutant
Adjuvant FOLFOX,
FOLFRI+bevacizumab
CAF 000V8 F IIIC Colon MSS, KRAS
mutant
Adjuvant
XELOX+bevacizuma,
modified to
Xeloda+bevacizumab
CAF 000UP M IV Colon TP53, and APC
mutant, MSS
Unknown
NF 000V8 F n/a Colon n/a n/a
NF 000VT M n/a Colon n/a n/a
CAF 000W8 M IVa Liver CHEK2 mutant,
MSS
Adjuvant Xeloda,
switched to FOLFOX
CAF 000UR M IIA
(primary)
Liver KRAS mutant Neoadjuvant
FOLFOX+Avastin
CAF 000UA F IIA
(primary)
Liver TP53, PI3KCA,
APC mutant
Chemo/XRT
CAF 000TZ M IV
(primary)
Liver MLH1, MSH2,
MSH6, PMS2
Adjuvant
FOLFOX+Avastin
CAF 000UJ M
Liver SMARCA4,
TSC2, TP53
mutant
adjuvant
FOLFOX+Avastin
CAF 000VF F
Liver n/a FOLFOX+Erbitux,
FOLFIRI+Avastin
NF 000VF F n/a Liver n/a n/a
NF 000VE M n/a Liver n/a n/a
NF 000VG M n/a Liver n/a n/a
62
Figure 2.2. Patient-derived CRC CAFs show heterogeneous morphology and expression of CAF markers. (A)
Representative brightfield images of CAF showing long, spindle-like, mesenchymal morphology. Three CAFs are
shown to highlight heterogeneous morphology across patients. (B) qPCR analysis of CAF-associated markers
fibronectin, aSMA, and vimentin was performed on primary cultured CAFs (isolated from CRC tissues of patients).
Expression was normalized to normal primary fibroblasts cell line CCD18Co. n=3.
63
CAFs Promote or Restrain Invasion in a Patient-Specific Manner
To better understand the functional heterogeneity, I used the CRC-Chip system to
interrogate the role of CAFs in CRC invasion (detailed discussion and data in Chapter 1). I first
examined the effects of CAF-derived secreted factors using conditioned media (CM) (Figure
2.3A). For the secreted factor analysis, I perfused CAF CM from 11 different patient-derived
CAF lines through the epithelial channel of the chip for the duration of the experiment (6 days).
As shown in Figure 2.3B, some CM resulted in increased HCT116 tumor cell invasion, some
resulted in decreased tumor cell invasion, and some had no effect. In order to better understand
the patient-specific phenotype, I measured the cytokines secreted in the CM from four different
patient-derived CAF lines (CAF000U8, US, W8, and UE) that significantly increased HCT116
invasion (Figure 2.3C). While some cytokines overlapped between the CAFs, there was
significant patient heterogeneity in the cytokine profiles (Figure 2.3D). When the heat map of
cytokine levels from all CAFs is overlayed with the on-chip invasion assay, some CAFs of
similar phenotypes cluster together, while others do not (Figure 2.3E).
64
Figure 2.3. CAF CM shows heterogeneous cytokine profiles and influences tumor cell invasion in a patient-
specific manner. (A) Schematic showing worm’s eye view of HCT116 CRC-Chip invasion assay with CAF
conditioned media. Caco2 C2BBe1 epithelial cells and HCT116 H2B-GFP tumor cell clusters in the epithelial channel
are exposed to secreted factors from CAF conditioned media that is continuously flowing through the top channel
throughout the experiment. HCT116 H2B-GFP tumor cells that invaded into the bottom channel and adhered to the
endothelial channel were quantitated via the on-chip invasion assay detailed in Chapter 1. (B) Conditioned media
from CAFs derived from four patients (N= 4 Chips for each patient-derived CAF) was flowed through the epithelial
channel for the duration of the experiment. Differences in HCT116 cell invasion ratio quantification is depicted. Data
are represented as boxplots and analyzed using 1-way ANOVA with multiple comparisons; *p<0.05; ***p<0.001;
****p<0.0001. (C) Raw images of cytokine array blots performed on conditioned media collected from patient-derived
CAFs. (D) Cytokine and growth factor expression was evaluated via cytokine arrays. Z-scores were determined and
the overlap of upregulated and downregulated cytokines (> or < 0.5 fold) across CAF lines is shown. N=2 lots of CM
for each CAF. (E) Cytokine and growth factor expression in the CAF conditioned media was evaluated via cytokine
arrays and hierarchical clustering was performed.
Aim 2 – To analyze the impact of cancer-associated fibroblast heterogeneity on CRC progression.
Aim 2.1 Rationale and Preliminary Studies
CAFs are the dominant stromal cell type in the TME and in-
fluence various aspects of tumorigenesis, including metasta-
sis and drug response, through secretion of cytokines and
remodeling of the ECM
62
. Tumor fibroblast percentages over
50% correlate with decreased survival rate in many cancer
types, including CRC
63
. Thus, targeting CAFs has gained mo-
mentum as a potential treatment option; however, recent pre-
clinical studies and clinical trials have shown mixed results,
primarily due to intra- and inter-patient CAF heterogeneity
64-
66
. The implications of CAF heterogeneity have only recently
come to light, as summarized in a 2020 Consensus State-
ment published in Nature Reviews Cancer
67
. In preliminary
studies, we used the chip platform to assess whether CAF
functional heterogeneity affects CRC cell invasion. We com-
pared conditioned media from 11 patient-derived CAFs and
1 patient-derived normal fibroblast (NF) line and identified dif-
ferential secretomes that either reduced or promoted tumor
cell invasion (Fig. 9). A divisive hierarchical algorithm was
employed to investigate potential patterns in CAF-secreted
cytokines
68
(Fig. 10). Cluster analysis showed that many CAF
lines with similar invasive phenotypes (Fig. 9) had also simi-
lar cytokine profiles. Currently, detailed and quantitative as-
sessment of CAFs in more physiologically relevant cancer
model systems is limited, especially in contexts where mechanical forces and other cell types, including endo-
thelial cells, exist, such as in our organ-on-chip platform. We have optimized the seeding of CAFs on chips to
monitor tumor cell invasion in the presence of CAFs (Fig. 11). We show that CAFs isolated from CRC patient
12737 promote tumor cell invasion while autologous normal fibroblasts isolated from an adjacent normal region
reduce invasion. Therefore, in this aim, we will investigate the impact of patient-isolated CAFs on CRC primary
tumor cell growth and invasive potential and perform mechanistic studies to define CAF subtypes (Fig. 12).
Aim 2.2 Research Design
I. Optimization of the CRC-on-Chip to co-
culture tumor and CAF cells. CAFs from our
biorepository or freshly isolated from patient
tissues will be used for these studies. Expres-
-smooth muscle ac-
-SMA), vimentin, fibroblast activated pro-
tein- FAP- ), and fibronectin) will be deter-
mined by qRT-PCR and immunofluorescence
prior to initiation of the experiments. The exper-
imental conditions that support favorable CAF
morphology and viability on chips are outlined
in Table 3. CAFs at low, medium, and high
densities (25, 50, 75%), reminiscent of CRC
clinical stromal percentages
63
, will be labeled
with CellTracker Orange, a non-toxic live cell
stain, and seeded in the top channel of chips
using a sandwich method –e.g., between two
different layers of ECM (bottom layer – Colla-
gen IV; top layer – Matrigel). The following day,
patient-matched normal colon and tumor or-
ganoids will be digested and seeded on top of
the ECM layer containing fibroblasts. Endothe-
lial cells will be seeded as described in Aim 1.
Figure 9. Heterogeneous CAF-induced invasion in
chips. Conditioned media from 11 CAF lines and 1 NF
line was continuously flowed in the top channel of the chip
for 6 days. HCT116 invasion ratio was calculated as de-
scribed in Fig. 2. *, P< 0.05; **, P<0.01; ***, P<0.001; ****,
P<0.0001.
Figure 10. CAF cytokine profiles. Cytokine profiles, featuring 105 cyto-
kines, were derived using the R&D Proteome Profiler Human Cytokine Ar-
ray platform. Raw signal was first corrected for background noise using ref-
erence and control spots and subsequently normalized using quantile-
quantile normalization. The median cytokine profile was generated through
aggregation across replicates and a divisive hierarchical algorithm was ap-
plied to the median normalized signal, in log10-space. The resulting rela-
tionships were superimposed with invasive phenotypes observed in Figure
9 for interpretation purposes.
CAF000W8
CAF000UE
CAF000U0
CAF000US
CAF000UM
CAF000UP
CAF000U8
CAF000UH
CAF000V8
CAF000UU
CAF000V8
Control
CAF000VT
CAF000U0
CAF000UP
CAF000V8
CAF000UU
CAF000UM
CAF000UH
CAF000U8
CAF000US
CAF000W8
CAF000UE
0
5
10
15
Conditioned Media
D6 Invasion Ratio
(Normalized to D0)
****
****
****
****
*
***
*
A. B.
CAF000UE
CAF000US
CAF000U8
CAF000W8
C.
D.
CAF Secreted Factors
E.
65
While the continual exposure of tumor cells to CAF secreted factors through
microfluidics is an exciting advancement over current methods, the physical interaction between
CAFs and tumor cells has been shown to be important for invasion (Gaggioli et al., 2007). I
utilized the method described in Chapter 1 to further study the physical interactions between
CAFs and tumor cells and the resulting tumor cell invasive phenotype (Figure 2.4A). Tumor cells
and HCT116 H2B-GFP cells showed substantial physical interactions, as initially reported in
Chapter 1 (Figure 1.11; Figure 2.4B). Like the CM results, CAFs resulted in heterogeneous
HCT116 tumor cell invasion on-chip, with some CAFs increasing and other CAFs decreasing
tumor cell invasion (Figure 2.4C). The presence of CAF000UE and CAF000W8 significantly
increased invasion, similar to the phenotype observed with the CM derived from the same CAF.
Interestingly, several CAFs produced different invasion phenotypes on-chip than with the CM
(specifically, CAF000UM and CAF000V8). The CM from these CAFs did not influence tumor cell
invasion, while the presence of these CAFs on-chip resulted in increased tumor cell invasion,
suggesting the physical interaction is important for the invasion phenotype. The ability to culture
CAFs and tumor cells in a 3D, microfluidic system is an exciting improvement over existing
methods and the results described here indicate the physical interaction between CAFs and
tumor cells requires further study.
66
Figure 2.4. Physical interactions between CAFs and tumor cells on-chip promote or restrain tumor cell
invasion in a patient-specific manner. (A.) Schematic showing worm’s eye view of HCT116 CRC-Chip invasion
assay with CAF co-cultured on-chip. Patient-derived CAFs were cultured on the top, epithelial channel and Caco2
C2BBe1 epithelial cells and HCT116 H2B-GFP tumor cell clusters were layered on top. HCT116 H2B-GFP tumor
cells that invaded into the bottom channel and adhered to the endothelial channel were quantitated via the on-chip
invasion assay detailed in Chapter 1. (B) Representative images of CAFs on day 0 (one day after tumor cell seeding)
illustrate heterocellular interactions on chip. CAFs were labeled with Cell Tracker Deep Red (purple). (C) CAFs were
labeled with Cell Tracker Deep Red and seeded in the top channel prior to epithelial cell and HCT116 cell seeding
(N=4 replicates for each patient-derived CAF) and the invasion ratio was quantified on day 6. Data are represented
as boxplots and analyzed using a 1-way ANOVA with multiple comparisons; *p<0.05; ***p<0.001; ****p<0.0001.
Gene Expression Analysis of CAFs Reveals Patient- and Organ- Specific Heterogeneity
To better determine the heterogeneous nature of the CAFs and understand the influence
of CAFs on CRC tumor invasion and progression, I performed RNAseq on 12 CAFs isolated
from the primary colon tumors (colon CAFs), 4 CAFs from colon tumors that have metastasized
to the liver (liver metastatic CAFs), and 5 normal fibroblasts (NFs) derived from adjacent tissue
(Table 2.1). I compared the gene expression between colon CAFs and liver metastatic CAFs
and found that CAFs clustered primarily based on location of the tumor from which the cells
were derived (Figure 2.5). In addition, I identified groups of colon CAFs and groups of liver
metastatic CAFs with similar gene expression profiles (Figure 2.5), respectively, suggesting
CAFs are a heterogeneous cell type and that there may be different subtypes of CAFs
represented within our samples.
No CAF Control
CAF000UP
CAF000VT
CAF000V8
CAF000W8
CAF000UE
CAF000UM
0
10
20
30
40
Fibroblast
D6 Invasion Ratio
(Normalized to D0)
***
****
****
***
*
Patient-derived CAFs
CAF000UP CAF000V8 CAF000UM CAF000VT
A.
B.
C.
67
Figure 2.5. CAFs show heterogeneous gene expression across the primary tumor and the liver metastatic
site. Heatmap of genes in patient-derived colon CAFs (teal) or metastatic liver CAFs (red). (n=12 and 4, 2 biological
replicates each).
Differentially expressed genes (DEG) were identified between the colon CAFs and the
liver metastatic CAFs and gene ontology (GO) was used to predict the functionality of these
genes. I observed differences in cell adhesion and migration between the two groups of CAFs
(Table 2.2), agreeing with previous work showing that liver metastatic associated fibroblasts
(MAFs) are more activated than CAFs derived from the primary colon tumor and show gene
expression signatures related to integrin signaling and ECM remodeling (Shen et al., 2020).
This preliminary and exploratory RNAseq analysis shows interesting differences between CAFs
derived from the primary colon tumors compared to the metastatic tumors. Future experiments
68
and analyses will be performed to better understand the differences between colon CAFs and
liver metastatic CAFs.
Table 2.2. Summary of GO terms (colon CAFs versus liver metastatic CAFs)
Direction
(Colon vs Liver)
Adjusted
P value
Number
of Genes
Pathways
Down regulated 3.83E-33 314 Anatomical structure morphogenesis
Down regulated 2.77E-27 276 Nervous system development
Down regulated 1.04E-24 361 Animal organ development
Down regulated 1.81E-23 405 Cell differentiation
Down regulated 8.11E-23 242 Movement of cell or subcellular component
Down regulated 8.11E-23 194 Biological adhesion
Down regulated 1.89E-22 407 Cellular developmental process
Down regulated 2.40E-22 192 Cell adhesion
Down regulated 2.55E-22 388 Regulation of biological quality
Down regulated 1.39E-21 286 Regulation of localization
Down regulated 1.22E-20 324 Regulation of multicellular organismal process
Down regulated 1.35E-20 278 Regulation of developmental process
Down regulated 1.67E-20 213 Locomotion
Down regulated 3.51E-20 231 Regulation of multicellular organismal
development
Down regulated 1.77E-19 180 Cell migration
Up regulated 4.84E-12 123 Cellular response to chemical stimulus
Up regulated 1.47E-11 103 Anatomical structure morphogenesis
Up regulated 4.11E-10 114 Regulation of multicellular organismal process
Up regulated 4.11E-10 55 Circulatory system development
Up regulated 8.19E-10 106 Response to external stimulus
Up regulated 8.19E-10 101 Cellular response to organic substance
Up regulated 9.86E-10 115 Response to organic substance
Up regulated 2.00E-09 59 Negative regulation of multicellular organismal
process
Up regulated 2.18E-09 136 Regulation of response to stimulus
Up regulated 5.55E-09 80 Tissue development
Up regulated 8.16E-09 106 Cell surface receptor signaling pathway
Up regulated 8.16E-09 64 Cell migration
Up regulated 1.54E-08 61 Positive regulation of developmental process
Up regulated 1.60E-08 50 Tube development
Up regulated 1.60E-08 79 Regulation of multicellular organismal
development
69
Colon CAFs Show Changes in Neurotransmitter Signaling Pathways
I then compared CAFs and NFs at each tumor site (colon and liver metastatic) and
identified DEGs within the two groups of CAFs. There were 99 genes enriched in colon CAFs
and 248 enriched in liver metastatic CAFs, compared to the site-specific NFs, with 12 genes
that overlapped between the two groups (Figure 2.6A). Of the 10 coding genes, several genes
of interest were identified, such as Wnt family member 2 (WNT2), gap junction beta 2 (GJB2),
and Ras interacting protein 1 (RASIP1) (Figure 2.6B), given the role of Wnt and Ras in CRC
(Nie et al., 2020). The enrichment of GABA type A receptor-associated protein (GABRAP) was
unexpected and I performed a GO pathway analysis using DEG between the colon CAFs and
NFs. Surprisingly, I found that many pathways related to neurotransmitter signaling, particularly
GABA receptor activation were dysregulated in the colon CAFs (Figure 2.7).
Figure 2.6. Enriched genes in CAFs. (A). Differentially expressed genes (DEG) were identified by comparing gene
expression of liver metastatic CAFs versus liver NF and colon CAFs versus colon NFs. The total number of enriched
genes and the 12 overlapping enriched genes is shown in the Venn Diagram. (B). The log 2 fold change of CAFs
versus NFs for the commonly enriched CAF genes is plotted for CAFs derived from the colon and the liver metastatic
tumor.
CAF Liver CAF Colon
n= 248 n= 99
239 87 12
WNT2 ATP8A2 GJB2 GABARAP RASIP1 PRSS2 EEF1A2 PRSS21 ARHGAP28
0
2
4
6
8
Gene
Log
2
Fold Change (CAF/NF)
Colon
Liver
A. B.
70
Figure 2.7. Pathway analysis indicate neurotransmitter signaling is changed within colon CAFs. Gene set
analysis (GSA) of differentially expressed genes (DEG) between colon CAFs versus NFs.
2.3 DISCUSSION
Tumor-stromal cell interactions are important to recapitulate in preclinical cancer model
systems, in particular, CAFs given their abundance in the TME. It is well documented that CAFs
can promote invasion (Karnoub et al., 2007; Orimo et al., 2005), which I demonstrated in the
CRC-Chip model. I show here that some CAFs have the ability to promote invasion, while
others restrain invasion or show no effect. It has been shown in a transwell experimental model
that some CRC CAFs promote CRC tumor cell migration, while others do not (Herrera et al.,
2013). on-chip invasion assay builds on this existing work by incorporating a more physiological
relevant system and the ability to distinguish physical versus secretome effects.
Moreover, it is now appreciated that CAFs are highly plastic and respond to diverse cues
from cancer cells and the TME (Biffi and Tuveson, 2021). Rather than existing in a terminally
differentiated state, CAFs can adapt to surrounding factors and interconvert between states,
thus influencing tumor cells in a diverse manner. An example of CAF plasticity was recently
highlighted in a pancreatic ductal adenocarcinoma (PDAC) study, whereby the authors identified
Number of differentially expressed genes
71
spatially and functionally distinct CAF subpopulations, termed inflammatory CAFs (iCAFs) and
myofibroblasts (myCAFs) (Ohlund et al., 2017). They concluded that myCAFs reside closer to
the tumor foci, while iCAFs are further away, and an intermediate subpopulation that expresses
both myCAF and iCAF markers was also detected. Further work by this group demonstrated
that the cell state transition between myCAFs and iCAFs was dependent on TGF-b and IL-
1/JAK/STAT signaling from the tumor cells (Biffi et al., 2019). A deeper understanding of CAF
plasticity in the context of tumorigenesis remains an important research challenge with
significant clinical implications. In order to begin to understand how the TME-tumor interactions
shape CAF plasticity, I show here that patient-derived CAFs are transcriptionally heterogeneous
and impact tumor intravasation in our OOC model in a patient-specific manner. My experimental
model is well-suited to delve into CAF plasticity and future work will investigate how CAF
subpopulations influence tumor progression.
Our preliminary studies suggest that pathways related to neurotransmitters are disrupted
in colon CAFs which, to my knowledge at the time of this writing, is the first study to highlight the
connection between neurotransmitter signaling in colon CAFs. Recently, the Lambrechts lab
performed a pan-cancer single-cell profiling of the tumor microenvironment and showed that the
GABA shunt pathway was changed between colon-specific fibroblasts and CAFs derived from
several cancers (Qian et al., 2020). While deriving more information from this study is
challenging as a direct comparison between colon fibroblasts and colon CAFs was not
performed, the authors specified that colon-specific fibroblasts expressed several markers
(KCNN3, P2RY1, and LY6H) responsible for neuronal signaling within the gastrointestinal tract
(Qian et al., 2020). In addition, a separate study compared colon CAFs and liver metastatic
CAFs by bulk RNAseq and showed that several pathways related to neurotransmitter release,
synapse formation, and nervous system development were significantly altered between the
groups of CAFs (Shen et al., 2020). Combined with my work here showing that
neurotransmitter-related pathways are changed within the colon CAFs and GABARAP is
72
upregulated within the CAFs, it is quite clear that the role of neurotransmitters in colon CAFs is
understudied and requires further work.
To conclude, it is evident that CAFs are important mediators of cell behavior in the TME
and although there are significant challenges to studying this cell type, understanding how these
cells interact with their surroundings (and systematically targeting these interactions) has the
potential to significantly impact patient care.
Limitations of the Study
The work described in this chapter is preliminary and exploratory, with the goal of
characterizing the Ellison Insitute biorepository of patient-derived CAFs to guide future
experiments studying tumor-CAF interactions and CAF plasticity in colon cancer progression.
The numbers of colon CAFs and liver CAFs were imbalanced, and the availability of NFs was
limited; future experiments will address these concerns.
It is important to note that the fibroblasts used in this work were cultured using traditional
2D cell culture methods prior to use on-chip or for sequencing studies. As such, the RNAseq
data needs to be validated with patient tissue. In addition, the treatment status of the patients
and the fibroblasts has not been considered for the results described in this chapter. Moving
forward, I will consider if the patient tissue was obtained prior or post treatment and I will
analyze the RNAseq data accordingly to understand if the treated CAFs show distinct
differences from the treatment naïve CAFs.
One limitation of the current chip model is the inability to perform time-lapse imaging
without sacrificing continual and controlled fluid flow and cyclic stretching. Future iterations of
this system may support the imaging demands required to capture the physical interactions
between tumor cells and CAFs and how this crosstalk drives or restrains intravasation.
73
2.4 FUTURE DIRECTIONS
The physical interaction between CAFs and tumor cells is of great interest; as such, the
on-chip invasion experiments will be expanded to include CAFs from more patients. In addition,
as the Mumenthaler lab expands our work on-chip utilizing patient-derived colon and CRC
organoids, we will introduce patient-matched CAFs to the system to study the effects on
invasion in a more physiologically-relevant iteration of the CRC-Chip. In addition, the RNAseq
analysis performed on the NFs and CAFs, while extremely exciting, is preliminary. We will be
delving more into the inter-patient heterogeneity and the differences between the CAFs derived
from the colon and the liver in future analyses.
To expand our understanding of CAF plasticity, I suggest a workflow that combines the
areas of biomimetic models, experimental tools, functional readouts, and mathematical models.
For instance, to better study the effect of CAFs on tumor metastasis, a biomimetic organ-on-
chip model could be used to accurately recreate the tissue-tissue interface of vessel structures
and cancerous tissue while tuning the model to include CAF vs no CAF scenarios (Strelez et al.,
2021). Combining an experimental technique, such as sc-RNAseq of CAFs, with the functional
readout of cancer cell intravasation on-chip, measured by live cell imaging, would yield powerful
insights into the phenotypic heterogeneity of CAFs and the effects of those phenotypic profiles
on invasion. A mathematical model could be developed from the sc-RNAseq and intravasation
data and used to generate predictions for how perturbations of the TME influence the
composition of CAF subtypes and subsequent tumor cell invasion behaviors. Together, these
scientific and technological advances offer unprecedented approaches to studying CAF
heterogeneity and, when combined with patient samples and large-scale clinical datasets, can
begin to connect the functional relevance of CAF heterogeneity to patient outcome.
74
2.5 METHODS
Patient-Derived Samples
Tissue resections were received from the USC Norris Comprehensive Cancer Center following
Institutional Review Board (IRB) approval (Protocol HS-06-00678; approval date 08-02-2019)
and patient consent. Tumor profiles, including known tumor mutations, sex, and treatment
information, are detailed in Table 2.1. Human primary fibroblasts were derived from CRC tumors
as described in the methods section of Chapter 1. Cells cultured on plastic tissue culture plates
were confirmed to be cancer-associated fibroblasts (CAFs) based on morphology and gene
expression signatures of vimentin (Vim), alpha smooth muscle actin (Acta2), and fibronectin
(Fn1), as measured via RT-qPCR (Figure 2.1). Primary CAFs were used for experiments
between passage 3 and passage 7.
RT-qPCR
To confirm the identity of CAFs, I performed RT-qPCR analysis on cultured primary cells.
Cellular RNA was extracted using RNAspin Mini RNA Isolation Kit (GE Healthcare; #25-0500-
71) and cDNA was reverse transcribed using iScript Reverse Transcription Supermix (Bio-Rad,
#1708841) following manufacturer’s instructions. The cDNA was then amplified using iScript
SYBR Green Master (Bio-Rad; #1708880). The sequences for PCR primers are listed in Table
2.3. Results were normalized to GAPDH expression for all experiments.
75
Table 2.3. Gene specific primers for qPCR, Related to Figure 2.1.
Gene Direction Sequence 5’ à 3’
Human GAPDH Forward TCTGGTAAAGTGGATATTGTTG
Reverse GATGGTGATGGGATTTCC
Human E-Cadherin Forward TTTGTACAGATGGGGTCTTGC
Reverse CAAGCCCACTTTTCATAGTTCC
Human EpCam Forward AATGTGTGTGCGTGGGA
Reverse TTCAAGATTGGTAAAGCCAGT
Human αSMA Forward CAATGGCTCTGGGCTCTGTAAG
Reverse TGTTCTATCGGGTACTTCAGGGTC
Human Fibronectin Forward TCCCTCGGAACATCAGAAAC
Reverse CAGTGGGAGACCTCGAGAAG
Human Vimentin Forward GAGAACTTTGCCGTTGAAGC
Reverse GCTTCCTGTAGGTGGCAATC
CAF Conditioned Media
CAFs were seeded into a 6-well plate and allowed to grow until they reached 70% confluency at
which point media was exchanged for fresh colon media. After 72 hours, the media was
collected, centrifuged at 900 rpm (200 rcf) for 5 minutes, and the supernatant was removed and
stored at -20
o
C. CAF conditioned media (CM) was diluted 1:1 with DMEM, 10% FBS, 1% Pen-
Strep for all experiments.
CAF Secretome Analysis
CAFs were seeded into a 6-well plate to collect CM as described above. When cells reached
70% confluency, one well of the 6-well plate was cultured in un-supplemented colon media
(ADMEM/F12) without FBS or Pen-Strep for 72 hours. This media was collected, spun down to
remove debris, aliquoted, and stored at -80
o
C. Frozen CM was thawed and analyzed using a
cytokine array (R&D Systems, #ARY022B) following manufacturer’s instructions. Analyses of
76
over- and under-expressed cytokines were performed in the R statistical environment (v4.0.2).
Z-scores were generated by gene-wise scaling and were assessed for overlap at thresholds of
0.5. Visualization of overlaps was facilitated by the VennDiagram package (v1.6.20) (Chen and
Boutros, 2011).
Tumor Cell Invasion Assay On-Chip
Detailed methods for the on-chip tumor cell invasion assay are described above in Chapter 1.
Gene expression profiling by RNAseq
Patient-derived CAFs and NFs (Table 2.1) were cultured using standard tissue culture
methods. RNA extraction, cDNA synthesis, sample quality assessment, cDNA library
preparation, and sample sequencing were performed by GENEWIZ from Azenta Life Sciences.
Paired end HiSeq 2500 (Illumina) sequencing was performed with 150 bp read length. Each
sample showed an average of 30 million reads. RNAseq pipeline was performed on the control
and treatment samples using RASflow (Zhang and Jonassen, 2020). FastQC (Babraham
Bioinformatics) was performed to check the quality of reads and alignment was performed using
HISAT2 to human genome GRCh38 (Hg38) (Kim et al., 2019).
Differential gene expression analysis was performed using DESeq2 for differential and
top differentially expressed genes were selected using fold change >2 and FDR adjusted p-
value <0.01(Love et al., 2014). Gene enrichment pathway analysis was performed using DAVID
(version 6.8) (https://david.ncifcrf.gov/) and GSEA (https://gsea-msigdb.org). R package
clusterProfiler was used to make plots for pathway analysis (Yu et al., 2012).
77
CHAPTER 3
Peristalsis-like deformations increase tumor cell intravasation
through GABAergic signaling
Contributions:
C. Strelez optimized all iterations of CRC-chip experiments, with the help of K. Ghaffarian. All
on-chip invasion assays were performed and analyzed by C. Strelez. FACS experiments were
performed by C. Shah. qPCRs were performed by C. Strelez and K. Ghaffarian and the
neurotransmitter PCR arrays were completed by K. Ghaffarian. Barrier function assays were
performed by C. Strelez. CTC RNAseq was analyzed by R. Sun. Mass spectrometry-based
metabolomics analyses were done by S. Chilakala, A.Y. Yoon, and J.E. Katz. Liver-Chip
extravasation studies were completed by C. Strelez.
3.1 INTRODUCTION
It is understood that cancer cells respond to mechanical cues present in the body (e.g.,
shear force from fluid flow) (Follain et al., 2020), however many of the mechanistic details
remain unknown as these processes are difficult to study. While microfluidic systems have
offered insights into the role of shear forces in tumor biology, few microfluidic OOCs incorporate
mechanical rhythmic deformations. A major advantage of our CRC-Chip model over other OOC
technologies is the ability to mimic peristalsis, a physiological process of muscle contraction and
relaxation that naturally occurs in the colon. Recent evidence suggests that mechanical forces
may accelerate cancer progression rather than act as a passive bystander (Ciasca et al., 2016)
and can alter the stromal milieu, which in turn influences tumor cell behavior (Huang et al.,
2013). In the context of CRC, it has been shown that increased pressure in the tumor promotes
tumor cell proliferation and adhesion through Src and FAK signaling (Basson et al., 2000). While
these reports investigated how various physical forces influenced cancer progression, another
78
recent study showed that CRC tumor cell lines representing degrees of metastatic potential
were distinguished based on cell deformability when exposed to shear stress, indicating there
may be prognostic value in understanding the mechanobiology of CRC progression (Armistead
et al., 2020). There remains a dearth of information on the importance of mechanical forces
during cancer progression; however, OOC technologies are primed to fill this scientific void.
In this section, I use the CRC-Chip, on-chip imaging, and effluent analysis described
extensively in Chapter 1 to study how CRC tumor cell invasion is influenced by mechanical,
peristalsis-like forces within the colon.
3.2 RESULTS
CRC Tumor Cell Invasion Increases in the Presence of Mechanical, Peristalsis-like
Forces
Harnessing the unique capabilities of the Human Emulation System, I performed on-chip
invasion experiments with or without mechanical peristalsis-like forces (10% strain; 0.2 Hz). In
Chapter 1, I showed that in the presence of peristalsis, HCT116 tumor cell invasion increased
significantly (Figure 1.11A and Figure 3.1A). In addition to the on-chip invasion assay, I
collected cells from the bottom effluent of stretched and not stretched HCT116 CRC-Chips on
day 6. As described in Chapter 1 (methods, Figure 1.10), these cells represent tumor cells
which have invaded from the epithelial channel to the endothelial channel and remain viable in
circulation (“CTC-like” cells). Stretched HCT116 CTC-like cells showed reduced E-Cadherin
expression and increased Vimentin expression (Figure 3.1B), suggesting these cells acquired
mesenchymal-like phenotypes associated with increased invasion. However, HCT116 tumor
cells from the top epithelial channel show decreased proliferation in the presence of stretching,
as shown by quantification of immunofluorescence staining of Ki67 in the HCT116 tumor cells in
the top channel (Figure 3.1C). qPCR analysis of HCT116 tumor cells harvested from the top
channel also indicated reduced Ki67 expression in the presence of mechanical stretching
(Figure 3.1D). Together, this data suggests that in the presence of peristalsis, HCT116 tumor
79
cells undergo the “go-or-grow” phenotypic switch, a hypothesis that proliferation and invasion
are mutually exclusive behaviors (Hatzikirou et al., 2012).
I expanded upon the initial HCT116 experiments to better understand how other CRC
cells responded to peristalsis. While Caco2 C2BBe1 Intestine chips showed no change in
invasion, HT29 tumor cells increased invasion in response to peristalsis (Figure 3.1A). In
addition, preliminary data from the patient-derived CRC organoid ORG000US showed a trend
toward increased invasion (N=3, p=0.1, experiments on-going), suggesting this phenomenon is
not exclusive to cell lines (Figure 3.1A). I also performed experiments with the HCT116 CRC-
Chip with human large intestine microvascular endothelial cells (HIMEC) replacing the human
umbilical vein endothelial cells (HUVEC) to confirm the increased invasive phenotype was not
mediated by HUVECs only. As shown in Figure 3.1A, HCT116 tumor cells showed a 2.8-fold
increase in invasion in the presence of peristalsis with HIMECs in the endothelial channel. The
2.8-fold increase in invasion with HIMECs is less than the 5.8-fold increase in invasion with
HUVECs, suggesting there may be differences in endothelial-tumor cell crosstalk or differences
in how endothelial cells respond to peristalsis. To conclude, while the Caco2 C2BBe1 intestine
epithelial cells and endothelial cells are likely responding to the mechanical forces, the CRC
tumor cells themselves were the target of our studies moving forward given the significant
changes in invasion phenotype due to stretching-like motions.
80
Figure 3.1. CRC tumor cells increase invasion and decrease proliferation in response to peristalsis. (A) Caco2
Intestine Chips or CRC-Chips (ORG000US, HT29, or HCT116) were prepared with HUVECs or HIMECs, as
indicated. Invasion was measured on day 6 via fluorescence microscopy and data is displayed as the fold change in
the invasion between stretched and unstretched conditions. The fold change for each condition was determined to be
significant by an unpaired student’s t-test (*p<0.05, **p<0.01; n=6 chips each condition; n=3 chips for ORG000US
experiments). (B) Cells were collected from the endothelial effluent of stretched and not stretched HCT116 CRC-
Chips. RT-qPCR indicated stretched chips had higher vimentin and lower E-cadherin expression. Results were
compared to GAPDH. Data are represented as mean ± SEM and analyzed using multiple unpaired t-tests;
***p<0.001, **p<0.01; n=3 chips). (C) The epithelial channel of the HCT116 CRC-Chips was stained for Ki67 in
stretched and not stretched conditions and the intensity of Ki67 in the HCT116 H2B-GFP tumor cell clusters was
quantitated using Harmony Image Analysis Software. Data is displayed as mean ± SEM and an unpaired student’s t-
test was performed to determine significance (**p<0.01; n=6 chips). (D) Cells were harvested from the top channel of
stretched and not stretched HCT116 CRC-Chips and HCT116 H2B-GFP tumor cells were isolated via FACS. RT-
qPCR indicated stretched chips had lower Ki67 expression. Results were compared to GAPDH. Data is displayed as
mean ± SEM and an unpaired student’s t-test was performed to determine significant (p=0.05; N=3 biological
replicates, with 3 pooled chips per biological replicate).
A. B.
C.
D.
Stretched Not Stretched
0.0
5.0×10
7
1.0×10
8
1.5×10
8
2.0×10
8
2.5×10
8
Ki67 Intensity
✱✱
0.0
0.1
0.2
0.3
0.4
0.5
Ki67 Expression Relative to GAPDH
Stretched Not Stretched
Caco2
C2BBe1
ORG000US Caco2 +
HT29
Caco2 +
HCT116
(HUVEC)
Caco2 +
HCT116
(HIMEC)
0
2
4
6
8
Invasion Fold Change
(Stretch / No Stretch)
**
*
**
Stretched Not Stretched
0
2
4
6
8
Expression Relative to GAPDH
E-Cadherin
Vimentin
✱✱✱
✱✱
81
To better understand the dynamics of the peristalsis-mediated invasion phenotype, I
monitored HCT116 tumor cell invasion every two days from day 0 to day 6 and performed
iterations of starting and stopping rhythmic stretching. As shown in Figure 3.2, the presence of
stretching (peristalsis) increased invasion, even when applied later in the experiment.
Conversely, removing the stretching (peristalsis) decreased invasion, seen most significantly
when removed on day 2 of the experiment (blue line). This data suggests the peristalsis-
mediated invasive phenotype involves cellular plasticity. Understanding the connection between
tumor cell behavior and mechanical forces requires further study.
Figure 3.2 CRC tumor cells respond dynamically to mechanical force perturbations. HCT116 CRC-Chips were
prepared (with HUVECs). Stretching was introduced or removed from chips at various timepoints through the
experiment. Invasion via fluorescence microscopy was measured every 2 days throughout the experiment. Data is
shown as mean ± SEM. N=4 chips for each condition.
I found the presence of mechanical stretching did not noticeably impact the formation of
tight junctions, as shown by immunofluorescence staining of ZO-1 in the epithelial and
0 2 4 6
-1
0
1
2
3
4
Days
Log Invasion Ratio
(Normalized to D0)
D0-D6 Stretch
D0-D6 No Stretch
D0-D2 Stretch, D2-D6 No Stretch
D0-D2 No Stretch, D2-D6 Stretch
D0-D4 Stretch, D4-D6 No Stretch
82
endothelial channel of HCT116-CRC-Chip (Figure3.3A). In addition, the presence of mechanical
stretching did not significantly change the ability of the HCT116-CRC-Chip epithelial cells to
form a stable intestinal barrier over the course of the experiments (Figure 3.3B). Alternative
experiments to test epithelial barrier disfunction (e.g., TEER, trans-epithelial electrical
resistance) are required to make conclusions regarding the role of peristalsis on tight junction
and functional barrier formation.
Figure 3.3 On-chip characterization of barrier function in response to peristalsis. (A) Representative confocal
immunofluorescent images of the epithelial (top) and endothelial (bottom) channels of the CRC-Chips in the presence
(left) or absence (right) of peristalsis stained for ZO-1 (gold) on day 6. DAPI (blue) labels the nuclei of Caco2 C2BBe1
cells in the epithelial channel and HUVECs in the endothelial channel. White arrows indicate HCT116 cells. Scale
bars represent 200 μm. Images are maximum projections that span a 15 μm Z-height in the epithelial channel and a
10 μm Z-height in the endothelial channel with a 5 μm step size. (B)The apparent permeability (P app) of the intestinal
epithelial cells in the top channel was not changed when HCT116 tumor cells were added to the CRC-Chips. The
concentration of inulin-FITC that diffused from the epithelial channel to the endothelial channel was used to calculate
P app (N=3 Chips). Data are represented as mean ± SEM and analyzed using a 2-way ANOVA; p>0.05.
CRC Tumor Cells Secrete GABA in Response to Peristalsis
To determine peristalsis-mediated changes in the CRC-Chip, I performed mass
spectrometry-based metabolomics on the effluent of stretched and not stretched CRC-Chips.
The differentially expressed metabolites, when mapped with Ingenuity Pathway Analysis,
indicated changes to the L-serin degradation pathway (p=6.97E-5), the phospholipase
metabolism pathway (p=6.02E-4), the glycine betaine degradation pathway (p=1.76E-3), the
pyruvate to lactate pathway (p=2.03E-2) and the all-trans-decaprenyl diphosphate biosynthesis
No Stretch
Epithelium
Endothelium
ZO-1 DAPI HCT116 H2B GFP
Stretch
0 2 4 6 8 10
0
1
2
3
Day
P
app
(cm/s) x 10
-6
Stretch
No Stretch
A.
B.
83
pathway (p=2.03E-2) in the epithelial channel. Surprisingly, the differentially expressed
metabolites in the top channel mapped significantly to the neurological disease state (p=2.55E-
4). To investigate this further, I re-analyzed the effluent from stretched and not stretched CRC-
Chips using a neurotransmitter-specific library. Several neurotransmitters were found to be
changed in the stretched conditions, as shown in Table 3.1
Table 3.1 Fold changes of neurotransmitter intensities in the epithelial channel effluent in
response to peristalsis as measured by mass spectrometry-based metabolomics.
Compound P-value
(Stretch versus Not
Stretch)
Fold Change
(Stretch versus Not
Stretch)
GABA 0.008 9.91
Tryptophan 0.021 13.1
Glutamine 0.002 2.16
Tyrosine 0.009 2.74
5-hydroxy
tryptophan
0.775 1.00
Epinephrine 0.06 1.61
Dopamine 0.066527 -1.305107
Serotonin 0.194497 -1.000889
Since GABA was highly significantly changed (p-value and fold change) between
stretched and not stretched conditions, I compared the levels of GABA in the epithelial effluent
in stretched versus not stretched Caco2 Intestine Chip and CRC-Chips with various CRC cells
(HCT116, HT29, ORG000US). As shown in Figure 3.4A, Caco2 C2BBe1 chips did not indicate
any changes in GABA levels in the epithelial effluent while CRC-Chips with either HT29 or
HCT116 tumor cells showed highly significant increases in GABA levels (1.5 fold for HT29 or 7
fold for HCT116 tumor cells) (Figure 3.4A). To further determine which cell type was responsible
84
for the secretion in GABA, I measured GABA levels in CRC-Chips with patient-derived CRC
organoids exclusively in the epithelial channel (without a Caco2 C2BBe1 epithelial cell layer).
The ORG000US chips produced significantly higher GABA concentrations in stretched
conditions, indicating the CRC tumor cells are secreting GABA in response to the peristalsis-like
motions (Figure 3.4A). To further confirm this hypothesis, I performed experiments with HCT116
CRC-Chips without HUVECs in the presence or absence of stretching. In the presence of
stretching, GABA levels were significantly increased in the HCT116 CRC-Chips, regardless of
whether the HUVECs were present (Figure 3.4A), indicating the endothelial cells are not
involved in the GABA secretion by the CRC tumor cells. While it is clear that the HUVECs are
required for CRC tumor cells to adhere to the endothelial compartment during the invasion
assay, as shown by the non-significant difference in invasion ratio between stretched and not
stretched HCT116 CRC-Chips without HUVECs (data displayed in Figure 1.11A), the increased
GABA secretion by the CRC tumor cells is not HUVEC-dependent. The interactions between
the endothelial cells and CRC tumor cells requires further study outside the scope of this
dissertation.
I then performed a neurotransmitter-specific PCR array on HCT116 tumor cells
harvested from the top channel of stretched and not stretched HCT116 CRC-Chips. Preliminary
analyses from the array show the expression of GAD1, an enzyme responsible for GABA
synthesis, was slightly increased in HCT116 from stretched chips (Figure 3.4B). In addition,
expression of several proteins involved in neurotransmitter release and secretion (SNPH, SYN2,
and SYT1) were increased in HCT116 from stretched chips (Figure 3.4B). Taken together,
these preliminary results suggest CRC tumor cells secrete GABA in response to peristalsis.
85
Figure 3.4. CRC tumor cells secrete GABA in response to mechanical, peristalsis-like stretching. (A) GABA
intensities in the epithelial effluent of stretched and not stretched chips, as measured by targeted mass spectrometry-
based metabolomics. Data is plotted as the fold change of the GABA intensity in stretched versus not stretched chips.
The fold change for each condition was determined to be significant by an unpaired student’s t-test (**p<0.01;
***p<0.001; ****p<0.0001; N=6 chips each for stretched and not stretched condition). (B) Gene expression of genes
related to GABA synthesis and GABA secretion were measured by a neurotransmitter-specific PCR array. HCT116
tumor cells of stretched and not stretched chips were harvested on day 6 from the epithelial channel and isolated via
FACs. Data is displayed as the gene expression fold change of stretch versus not stretch conditions (N=3 biological
replicates with 3 pooled chip per biological replicate). Expression was normalized to GUSB, the housekeeping gene
with the lowest standard deviation across all replicates.
Peristalsis-Mediated CRC Tumor Cell Invasion is GABA RA-Mediated
I hypothesized that GABA signaling pathways could be involved in the invasion
phenotype, in response to the highly GABAergic environment. Two GABA receptors (the
ionotropic GABA receptor A and the metabotropic receptor B) respond to exogenous GABA and
have been implicated in cancer progression (Jiang et al., 2020). I performed a neurotransmitter-
specific PCR array to identify changes in gene expression in GABA receptor genes. My initial
results showed that HCT116 tumor cell GABBR1 expression (a subunit of GABA RB) did not
change in response to peristalsis, while GABRB2 and GABRQ (two subunits of GABA RA)
increased >2-fold (Figure 3.5A). I confirmed expression by immunofluorescence staining of
GABA RA and GABA RB in the bottom channel of CRC-Chips. In stretched conditions, invaded
HCT116 stained positive for GABA RA and negative for GABA RB, suggesting CRC tumor cells
increase expression of GABA RA in response to peristalsis (Figure 3.5B).
A. B.
GAD1 SNPH SYN2 SYT1
0.0
0.5
1.0
1.5
2.0
Gene Expression Fold Change
(Stretch / Not Stretch)
GABA
Synthesis
GABA
Secretion
Caco2 C2BBe1 Caco2 +
HT29
ORG000US Caco2 +
HCT116
Caco2 +
HCT116,
No HUVECs
0
2
4
6
8
Fold Change GABA Intensity
(Stretch/Not Stretch)
****
**
***
***
86
I then looked at GABA receptor agonists and antagonists to confirm the role of the
GABA RA in the invasion phenotype (Figure 3.5C). When the GABA RA antagonist bicuculline
was introduced to the top channel of the stretched chips, invasion was significantly reduced,
indicating GABA RA signaling was involved in the peristalsis-mediated invasion. Conversely,
adding exogenous GABA to the not stretched chips increased invasion significantly, suggesting
GABA alone is sufficient to increase invasion in the CRC-Chip model system. Interestingly, the
GABA RA agonist Muscimol was not sufficient to increase invasion significantly in the stretched
chips. Muscimol binds to GABA RA but cannot be metabolized by cells unlike GABA (Lüddens et
al., 1990; Yazulla and Brecha, 1981), suggesting that there may also be a role of GABA
metabolism in the increased invasive capabilities of the CRC tumor cells in the GABAergic,
peristaltic environment. Current experiments are ongoing to better understand if CRC tumor
cells transport GABA into the cell to use as an energy source or for further downstream
signaling that may increase tumor cell invasion.
I also performed similar experiments with agonists or antagonists targeting GABA RB
(Figure 3.5C). Peristalsis-induced invasion did not change in the presence of the GABA RB
antagonist CGP 54626. In addition, the GABA RB agonist R-baclofen did not increase invasion
in the not stretched conditions. Taken together, these results indicate that GABA RB is not
involved in the increased invasive capabilities of the HCT116 tumor cells in a GABAergic
environment.
87
Figure 3.5. CRC tumor cells respond to peristalsis through GABA RA signaling. (A) Gene expression of genes
related to GABA R A and R B were measured by a neurotransmitter-specific PCR array. HCT116 tumor cells of
stretched and not stretched chips were harvested on day 6 from the epithelial channel and isolated via FACs. Data is
displayed as the gene expression fold change of stretch versus not stretch conditions (N=3 biological replicates with
3 pooled chip per biological replicate). Expression was normalized to GUSB, the housekeeping gene with the lowest
standard deviation across all replicates. (B) Representative confocal immunofluorescent images of the endothelial
(bottom) channel of the CRC-Chips stained for GABA R A (red, left) or GABA R B (red, right) on day 6. Invaded
HCT116 H2B-GFP stain positive for GABA R A and negative for GABA R B, while HUVECs stain strongly for GABA R B.
Scale bars represent 200 μm. Images are maximum projections that span a 15 μm Z-height in the endothelial
channel with a 5 μm step size. (C) HCT116 CRC-Chips were prepared and GABA receptor agonists or antagonists
were introduced to the epithelial channel on day 0 as indicated. Invasion was measured on day 6 via fluorescent
microscopy. Data is shown as mean ± SEM and a one-way ANOVA was performed to determine significance
(**p<0.01, ***p<0.001, ****p<0.0001; N=5-6 chips per condition).
No Stretch No Stretch
+ GABA
No Stretch
+ Muscimol
No Stretch
+ R-Baclofen
Stretch Stretch
+ Bicuculline
Stretch
+ CGP 54626
0
5
10
15
Invasion Ratio Normalized to D0
✱✱✱
✱✱
ns
✱✱✱✱
ns
ns
GABA R
A
Agonist
GABA R
B Agonist
Antagonist
Antagonist
Agonist
200 µm 200 µm
GABA R
A
GABA R
B
Invaded HCT116 HUVEC (GABA R
B
+)
A. B.
C.
GABRB2 GABRQ GABBR1 GAD1 SNPH SYN2 SYT1
0
2
4
6
8
Gene Expression Fold Change
(Stretch / Not Stretch)
Preliminary Array Data
GABA R
A
GABA R
B
GABA
Synthesis
GABA
Secretion
88
Invaded, Circulating CRC Tumor Cells from Stretched CRC-Chips Show Increased
Adhesive Properties and can Colonize the Liver
I performed RNAseq on invaded, circulating tumor cells collected from the endothelial
effluent of the stretched and not stretched CRC-Chips on day 6 (see Figure 1.10D). Not only did
the stretched tumor cells show significant changes to GABAergic-related pathways (Figure
3.6A), additional pathways related to cell migration, integrin binding, and cell-cell adhesion were
significantly changed (Figure 3.6A). In addition, ICAM1 expression was significantly increased in
stretched CTCs (Figure 3.6B).
Based on the changes in adhesion properties in stretched CTCs and the role of
adhesion molecules in metastasis, I investigated if peristalsis also increases the ability of the
CRC tumor cells to metastasize to the liver. I performed the liver extravasation assay described
in detail in Chapter 1 (Figures 1.12-1.14). Briefly, CRC tumor cells are collected from the
HCT116 CRC-Chip and flown through the endothelial channel of a healthy human liver chip
comprised of LSECs, stellate cells, and Kupffer cells in the bottom, non-parenchymal channel
and hepatocytes in the top, parenchymal channel. Chips are imaged throughout the experiment
and CRC tumor cells (HCT116 H2B-GFP) are quantitated via image analysis software. CTCs
from stretched chips adhered to the non-parenchymal channel in significantly higher numbers
than CTCs from not-stretched chips, supporting the RNAseq data (Figure 3.6C). In addition,
CTCs from stretched CRC-Chips extravasated into the hepatocyte cell compartment more often
(hepatocyte colonization in 3/4 liver chips) than CTCs from not stretched chips (hepatocyte
colonization in 1/4 liver chips), as shown in Figure 3.6D. Taken together, these preliminary
results suggest that peristalsis may also be driving a more aggressive and metastatic CRC
tumor.
89
Figure 3.6 Peristalsis induces adhesion changes in CRC CTCs, resulting in increased extravasation in a
healthy human liver model. (A) CTCs were collected from stretched and not stretched HCT116 CRC-Chips and
RNAseq was performed. GO Pathway analysis was performed on a subset of genes with either a 2-fold difference
between stretched and not stretched CTCs or an FDR-adjusted p-value <0.1. N=2 biological replicates with 3 pooled
chips in each replicate. (B) Differentially expressed genes between stretched and not stretched CTCs were
determined based on a log2 fold change and a p-value. Genes are shown that had a q-value of less than 0.05. (C)
Stretched CTCs were flown through the bottom non-parenchymal channel of healthy human Liver-Chips. Chips were
imaged using confocal fluorescent microscopy on day 0 (the day after CTC flow-through) and quantified based on
GFP intensity. The number of GFP+ HCT116 CTCs in the non-parenchymal channel of the Liver Chips is plotted as
the mean ± SEM and an unpaired student t-test was performed to determine significance (**p<0.01; N=4 chips per
condition). (D) Liver Chips were imaged via confocal fluorescent microscopy on day 14 and GFP+ HCT116 CTCs in
the top, parenchymal channel of the healthy human Liver Chip was quantified. Data is displayed as the percent of
Liver Chips with extravasated CTCs (N=4 chips per condition).
Stretched CTC Not Stretched CTC
0
20
40
60
Number of HCT116 in
Endothelial Channel Channel
✱✱
Stretched CTC Not Stretched CTC
0.0
0.2
0.4
0.6
0.8
Percent of Liver Chips with HCT116 Extravasation
(HCT116 Present in Hepatocyte Channel on Day 14)
A. B.
C.
D.
0.00 0.01 0.02 0.03
Positive regulation
of phosphorylation
Response to drug
Negative regulation
of apoptotic regulation
Positive regulation
of cell adhesion
Wnt signaling
Cell-cell adhesion
Regulation
of cell migration
Integrin binding
Cell adhesion
molecule binding
Regulation of GABAergic
synaptic transmission
P-value
GO Pathway
90
3.3 DISCUSSION
Neurotransmitters have been implicated in a variety of cancers (Jiang et al., 2020). Here,
I show for the first time that CRC tumor cells secrete GABA in a peristaltic environment. The
role of GABA within colon cancer has been understudied, even though GABA is well understood
to be a major mediator in the function of the GI tract (Hyland and Cryan, 2010). GABA levels
have been shown to be higher in colon cancer than in normal colon tissue (Kleinrok et al., 1998)
and increased GAD1 levels have been correlated with worse survival in patients with stage
T3/T4 colorectal cancer (Yan et al., 2016), however studies trying to determine further
mechanisms have been conflicting. Some research suggests inhibiting GABA RB in colon
cancer reduces proliferation, invasion, and metastasis (Shu et al., 2016; Thaker et al., 2005).
Conversely, some studies found that expression of several subunits of GABA RA was increased
in CRC tissue and predictive of worse patient outcome (Niu et al., 2020; Wu et al., 2020; Yan et
al., 2020). The discrepancy between in vitro and in vivo data may, in part, be due to the lack of
physiological relevance within traditional in vitro experiments. I show here that peristalsis-like
mechanical forces drives the GABA secretion by tumor cells, however mechanical forces are
not modelled in many in vitro model systems.
Even within the OOC field, there are relatively few microfluidic Organ Chips that
incorporate mechanical rhythmic deformations. A non-small-cell lung cancer (NSCLC) Organ
Chip model demonstrated that rhythmic mechanical forces mimicking breathing decreased
NSCLC cell invasive behavior (Hassell et al., 2017). However, in this CRC model, I show that
peristalsis-like motions increase tumor cell invasion. This discrepancy may be explained by
different organs responding differentially to biophysical cues. Recent evidence suggests that
these mechanical forces may accelerate cancer progression rather than act as a passive
bystander (Ciasca et al., 2016) and can alter the stromal milieu, which in turn influences tumor
cell behavior (Huang et al., 2013) and is of particular interest moving forward, given the role of
neurotransmitter signaling in the CAFs outlined in Chapter 2. In the context of CRC, it has been
91
shown that increased pressure in the tumor promotes tumor cell proliferation and adhesion
through Src and FAK signaling (Basson et al., 2000). Stretching induced changes to adhesion
properties in our CTC-like cells, including increasing ICAM1 expression. A group recently
identified ICAM1 as an important mediator for CTC cluster formation and trans-endothelial
migration in breast cancer metastases (Taftaf et al., 2021). In addition, another group found that
“mechanically fit” CTCs with high deformability and high adhesion were able to interact with
immune cells and were more metastatic (Osmulski et al., 2021). While further experiments are
needed, the data presented here and in the literature suggests that the greater adhesion
observed in our model under peristaltic conditions, both by invading tumor cells adhering more
to the endothelial cell layer and by genetic changes in CTC-like cells, may be representative of
a more aggressive CRC cell.
While these reports investigated how various physical forces influenced cancer
progression, another recent study showed that CRC tumor cell lines representing degrees of
metastatic potential were distinguished based on cell deformability when exposed to shear
stress, indicating there may be prognostic value in understanding the mechanobiology of CRC
progression (Armistead et al., 2020). In addition, mechanically sensitive calcium ion channels
such as Piezos are important in mechanotransduction and Piezo1 is up-regulated in CRC and
correlated with poor prognosis in CRC patients (Sun et al., 2020; Wu et al., 2020; Yan et al.,
2020).
It is quite clear that more work needs to be done to understand how CRC tumor cells
change in response to mechanical forces and due to a GABAergic environment. While there
remains a dearth of information on the importance of mechanical forces and neurotransmitters
during cancer progression, due to inadequate model systems, the Organ Chip technologies are
primed to fill this scientific void.
92
Limitations of the Study
While I suggest this CRC-Chip is capable of establishing an intact barrier based on ZO-1
expression and permeability assays, an important caveat to this finding is the lack of
transepithelial/transendothelial electrical resistance (TEER) measurements of the tight junction
integrity.
The work described in this thesis is on-going and the data represent my hypotheses at the
time of submission. I acknowledge that more work needs to be done to understand the GABA-
related increased invasion and work in ongoing. In addition, the liver extravasation studies
suggesting stretched CRC CTCs extravasate and colonize a healthy liver more than not
stretched CRC CTCs are preliminary and need to be repeated within our liver chip model as
well as within an animal model.
3.4 FUTURE DIRECTIONS
This project will be continued to understand the increased CRC tumor cell invasion with the
hope of identifying potential therapeutic targets. A short-term goal for this project is working
toward a publication on the mechanism connecting mechanical forces, GABA, and CRC tumor
cell intravasation. To complete this goal, I are working on several follow-up experiments outlined
below.
Ongoing Experiments:
1. Perform GABA RA and RB knockdown experiments in CRC tumor cells to confirm
the role of GABA receptors in the peristalsis-mediated invasion phenotype. The
GABA receptor agonist and antagonist experiments described above implicated GABA
RA in the peristalsis-mediated invasive phenotype. I am currently working on creating
GABA RA and RB shRNA knockdowns in HCT116 tumor cells. I will then perform
stretched versus not stretched CRC-Chip experiments to see the invasion phenotype. I
93
hypothesize that the GABA RA knockdown will negate the peristalsis-mediated increased
invasion, while the GABA RB knockdown will not.
2. Measure GABA metabolic fluxes through
13
C magnetic resonance spectroscopy. I
am currently optimizing experimental conditions and will introduce
13
C-labeled GABA
into stretched and not stretched chips. Effluent media will be collected, cells will be
harvested from the epithelial channel, and the amount of
13
C-labeled GABA in the media
and within the cells will be measured. This experiment will answer questions related to
how the CRC tumor cells are metabolizing GABA and how mechanical stretching
impacts this GABA metabolism.
3. Understand how GABA metabolism-related enzymes are changed within the
stretched CRC tumor cells. I am currently trying to understand if stretching influences
ABAT and ALDH5a1 protein or gene expression via Western Blots and qPCR. If these
proteins or genes are upregulated within stretched chips, this will support our preliminary
hypothesis that CRC tumor cells metabolize GABA in a peristaltic, GABAergic
environment.
4. Understand expression of GABAergic genes in patient tumors. I am currently
working with a collaborator, Dr. Heinz-Josef Lenz, to interrogate if GABAergic genes
(GABA receptors, enzymes related to GABA metabolism, and GABA transporters) are
associated with outcome in a CRC patient dataset.
5. Interrogate the connection between mechanical forces and GABAergic changes
within CRC tumor cells. I do not currently understand how mechanical forces result in
increased GABA RA expression and GABA secretion within CRC tumor cells. I will
perform qPCR analyses on CRC tumor cells from the stretched and not stretched chips
to identify if the mechano-sensitive channel Piezo1 is upregulated in stretched CRC-
Chips. If so, I will perform knockdown experiments using shRNA targeting Piezo1 and
measure on-chip invasion and GABA in the effluent.
94
3.5 METHODS
Cell Culture
Cell lines and patient-derived samples were purchased or isolated and maintained as described
in Chapter 1. Human large intestine microvascular endothelial cells (HIMECS; Cell Systems,
#ACBRI 666) were maintained in Endothelial Cell Growth Medium MV2 (Promo Cell, C-22121),
containing 5 ng/mL recombinant human epidermal growth factor (rhEGF), 10 ng/mL
recombinant human basic fibroblast growth factor (rhbFGF), 20 ng/mL long R3 insulin-like
growth factor (R3-IGF), 0.5 ng/mL recombinant human vascular endothelial growth factor 165
(rhVRGF), 1 µg/mL ascorbic acid, 0.2 µg/mL hydrocortisone, 2% fetal bovine serum, and
1:1,000 (vol/vol) Primocin (InvivoGen, #ant-pm).
Microfluidic Organ-Chip Design and Culture
Chips were prepared as described in Chapter 1. For Caco2 + HCT116 experiments with
HIMECs, the top channel was coated with ECM as described in Chapter 1 and the bottom
channel was coated with collagen IV (200 µg/mL; Sigma Aldrich, #C5533) and fibronectin (30
µg/mL; Gibco, #22010-018). HIMECs were seeded in the bottom channel (1.2x10
5
cells in 20
μL; 7x10
5
cells cm
-2
) as described above in Chapter 1. Caco2 and CRC tumor cells were
seeded as described in Chapter 1. Fully supplemented Endothelial Cell Growth Medium MV2
was perfused through the bottom channel.
Tumor Cell Invasion Assay On-Chip
Detailed methods for the on-chip tumor cell invasion assay are described above in Chapter 1.
Circulating tumor cells were collected from the endothelial effluent as described in Chapter 1.
Isolation of CRC Tumor Cells From CRC-Chips
Cells were harvested from the top channel of the CRC-Chips using TrypLE incubations and the
resulting heterogeneous cell population was passed through a 50 µm strainer to obtain a single
cell suspension. Cells were resuspended in wash buffer (PBS + 4% FBS) and Fluorescence-
95
Activated Cell Sorting (FACS) (BD FACS Melody cytometer) was used to isolate HCT116 H2B-
GFP cells from Caco2-C2bbe1 cells for subsequent downstream analysis. Once the HCT116
H2B-GFP population had been identified based on GFP signal, the cytometer sorted these cells
based on a purity sort and snap frozen for downstream analyses.
RT-qPCR
To interrogate epithelial or mesenchymal marker expression or Ki67 expression, I performed
RT-qPCR on HCT116 harvested from the top epithelial channel and isolated via FACS. RT-
qPCR was performed as described in chapter 1. The sequences for PCR primers are listed in
Table 3.2. Results were normalized to GAPDH expression for all experiments.
Table 3.2 Gene specific primers for qPCR. Related to Figure 3.1.
Gene Direction Sequence 5’ à 3’
Human GAPDH Forward TCTGGTAAAGTGGATATTGTTG
Reverse GATGGTGATGGGATTTCC
Human E-Cadherin Forward TTTGTACAGATGGGGTCTTGC
Reverse CAAGCCCACTTTTCATAGTTCC
Human Vimentin Forward GAGAACTTTGCCGTTGAAGC
Reverse GCTTCCTGTAGGTGGCAATC
Human Ki67 Forward CTTTGGGTGCGACTTGACG
Reverse GTCGACCCCGCTCCTTTT
On-Chip Permeability Assay
Apparent permeability was investigated as described in Chapter 1.
Neurotransmitter PCR Array
I used a TaqMan
TM
Human Neurotransmitter PCR Array (ThermoFisher Scientific #4414094) to
detect the expression of GABA-related genes in HCT116 harvested from the top epithelial
channel and isolated via FACS. The array includes 84 genes known to be involved in
96
neurotransmitter synthesis, transport, and degradation, including GABAergic genes. RNA was
isolated and cDNA was transcribed as previously described in Chapter 1.
CRC-Chip Immunofluorescence
CRC Organ Chips were fixed and stained as described in Chapter 1. The primary antibodies
used for the CRC-Chip studies were rabbit anti-Ki67 (1:100; Abcam #ab15580), mouse anti-ZO-
1 (Invitrogen #339194), mouse anti-GABA A Receptor β2/3 (1:100; EMD Millipore Sigma #05-
474), and mouse anti-GABA B Receptor 1 (1:100; Abcam #ab55051).
Tumor Cell Invasion Assay On-Chip
Invasion was monitored and quantitated as described in Chapter 1. For GABA receptor agonist
and antagonist studies, the following drugs were perfused through the top channel starting on
day 0. GABA (1 µg/mL; Sigma Aldrich #A2129), 1 (S), 9 (R)-(-)-Bicuculline methiodide (GABA
RA antagonist) (10 µM; Sigma Aldrich # 14343), Muscimol (GABA RA agonist) (1 µg/mL; Tocris
#0289), CGP 54626 (GABA RB antagonist) (1 µM; Tocris #1088), and R-Baclofen (GABA RB
agonist) (10 µM; Tocris #0796). Inlet media with the GABA agonists or antagonists was
refreshed on day 3 of the experiment.
Mass Spectrometry-Based Metabolomics of CRC-Chip
Metabolite extraction and LC-MS/MS data acquisition was performed as described in Chapter 1.
Data was analyzed as described in Chapter 1 and compounds were identified using our in-
house neurotransmitter metabolites library using m/z values and retention time.
CTC RNAseq
Cells were collected from the endothelial effluent of stretched and not stretched CRC-Chips as
described in Chapter 1. Cells from 2 chips were pooled together to obtain enough cells for RNA
sequencing. RNA extraction, cDNA synthesis, sample quality assessment, cDNA library
preparation, and sample sequencing were performed by GENEWIZ from Azenta Life Sciences.
Samples were sequenced on HiSeq 2500 (Illumina) rapid run flow cells. Read length was 150
bp. Reads were mapped on reference human genome GRch38. Quality control was performed
97
using FastQC software. Gene expression quantification was performed using RSEM algorithms.
Gene sets were identified from a subset of genes that either had a 2-fold difference between
stretched and non-stretched or a false discovery rate (FDR)-adjusted p-value <0.1. Here, the
alpha level was increased to 0.1 from 0.05 to enable discovery with small sample sizes.
98
CHAPTER 4
Conclusions
This project evolved into understanding the role of neurotransmitters within colon cancer,
as I discovered altered neurotransmitter-related pathways, specifically GABA, in colon and liver
metastatic CAFs (outlined in Chapter 2) and the role of GABA in the peristaltic CRC (outlined in
Chapter 3). My long-term and ambitious goal for this project would be to target the GABA
pathway within the tumor cells to block the increased invasive properties of the tumor cells in a
GABAergic environment. While this may be challenging due to the role GABA plays within the
nervous system, future experiments may elucidate a related mechanism that is more feasible to
target.
Without studying the effect of peristalsis on CRC tumor invasion in this unique CRC-Chip
model system, the significance of the GABAergic changes in colon CAFs might have been
overlooked. This leads to further questions about how other microenvironmental factors
described in this dissertation (fluidics, stretching, endothelial cells, etc.) change the phenotypes
of the CAFs. It has been shown that fibroblasts respond to mechanical stretching by becoming
more activated, remodeling fibronectin, and altering cancer cell migration (Ao et al., 2015).
Therefore, investigating how CAFs respond to physical forces and how CAFs respond to a
GABAergic TME is an intriguing and novel next step for the project described in this dissertation
and for our CRC-Chip model system. My dream would be to develop drugs that target the
CAFs. If there is cross talk between the CAFs and tumor cells in this GABAergic environment,
targeting GABA-related pathways in the CAFs and the tumor cells may produce a better
outcome for patients.
For both the CAF and peristalsis aspects of this dissertation, it was crucial to develop a
physiologically relevant model system that was capable of tuning aspects of the TME to study
novel tumor biology. I hope the work presented in this thesis justifies the use of complex in vitro
99
model systems to uncover new biology, rather than just in the toxicology or drug discovery
space. There are many aspects of the TME that need to be incorporated into this model system
(immune cells, microbiome, nervous system) and I am excited for future models of this CRC-
Chip. To conclude, the significance of this dissertation is the intersection of model development
and mechanistic tumor biology, with the goal of identifying new therapeutic targets by
recapitulating and tuning the TME in vitro.
100
REFERENCES
Aleman, J., and Skardal, A. (2019). A multi-site metastasis-on-a-chip microphysiological system
for assessing metastatic preference of cancer cells. Biotechnol Bioeng 116, 936-944.
10.1002/bit.26871.
Amos, S.E., and Choi, Y.S. (2021). The Cancer Microenvironment: Mechanical Challenges of the
Metastatic Cascade. Front Bioeng Biotechnol 9, 625859. 10.3389/fbioe.2021.625859.
Ao, M., Brewer, B.M., Yang, L., Franco Coronel, O.E., Hayward, S.W., Webb, D.J., and Li, D.
(2015). Stretching fibroblasts remodels fibronectin and alters cancer cell migration. Sci Rep 5,
8334. 10.1038/srep08334.
Armistead, F.J., Gala De Pablo, J., Gadêlha, H., Peyman, S.A., and Evans, S.D. (2020). Physical
Biomarkers of Disease Progression: On-Chip Monitoring of Changes in Mechanobiology of
Colorectal Cancer Cells. Sci Rep 10, 3254. 10.1038/s41598-020-59952-x.
Bai, J., Tu, T.Y., Kim, C., Thiery, J.P., and Kamm, R.D. (2015). Identification of drugs as single
agents or in combination to prevent carcinoma dissemination in a microfluidic 3D environment.
Oncotarget 6, 36603-36614. 10.18632/oncotarget.5464.
Basson, M.D. (2007). Effects of repetitive deformation on intestinal epithelial cells.
Inflammopharmacology 15, 109-114. 10.1007/s10787-007-1562-8.
Basson, M.D., Yu, C.F., Herden-Kirchoff, O., Ellermeier, M., Sanders, M.A., Merrell, R.C., and
Sumpio, B.E. (2000). Effects of increased ambient pressure on colon cancer cell adhesion. J Cell
Biochem 78, 47-61. 10.1002/(sici)1097-4644(20000701)78:1<47::aid-jcb5>3.0.co;2-m.
Batista, S., Gregorio, A.C., Hanada Otake, A., Couto, N., and Costa-Silva, B. (2019). The
Gastrointestinal Tumor Microenvironment: An Updated Biological and Clinical Perspective. J
Oncol 2019, 6240505. 10.1155/2019/6240505.
Bhatia, S.N., and Ingber, D.E. (2014). Microfluidic organs-on-chips. Nat Biotechnol 32, 760-772.
10.1038/nbt.2989.
Biffi, G., Oni, T.E., Spielman, B., Hao, Y., Elyada, E., Park, Y., Preall, J., and Tuveson, D.A. (2019).
IL1-Induced JAK/STAT Signaling Is Antagonized by TGFβ to Shape CAF Heterogeneity in
Pancreatic Ductal Adenocarcinoma. Cancer Discov 9, 282-301. 10.1158/2159-8290.CD-18-0710.
Biffi, G., and Tuveson, D.A. (2021). Diversity and Biology of Cancer-Associated Fibroblasts.
Physiol Rev 101, 147-176. 10.1152/physrev.00048.2019.
Brabletz, T., Kalluri, R., Nieto, M.A., and Weinberg, R.A. (2018). EMT in cancer. Nat Rev Cancer
18, 128-134. 10.1038/nrc.2017.118.
Bürtin, F., Mullins, C.S., and Linnebacher, M. (2020). Mouse models of colorectal cancer: Past,
present and future perspectives. World J Gastroenterol 26, 1394-1426.
10.3748/wjg.v26.i13.1394.
Caballero, D., Kaushik, S., Correlo, V.M., Oliveira, J.M., Reis, R.L., and Kundu, S.C. (2017). Organ-
on-chip models of cancer metastasis for future personalized medicine: From chip to the patient.
Biomaterials 149, 98-115. 10.1016/j.biomaterials.2017.10.005.
Carvalho, M.R., Barata, D., Teixeira, L.M., Giselbrecht, S., Reis, R.L., Oliveira, J.M.,
Truckenmuller, R., and Habibovic, P. (2019). Colorectal tumor-on-a-chip system: A 3D tool for
precision onco-nanomedicine. Sci Adv 5, eaaw1317. 10.1126/sciadv.aaw1317.
101
Chen, H., and Boutros, P.C. (2011). VennDiagram: a package for the generation of highly-
customizable Venn and Euler diagrams in R. BMC Bioinformatics 12, 35. 10.1186/1471-2105-12-
35.
Choi, H., and Moon, A. (2018). Crosstalk between cancer cells and endothelial cells: implications
for tumor progression and intervention. Arch Pharm Res 41, 711-724. 10.1007/s12272-018-
1051-1.
Ciasca, G., Papi, M., Minelli, E., Palmieri, V., and De Spirito, M. (2016). Changes in cellular
mechanical properties during onset or progression of colorectal cancer. World J Gastroenterol
22, 7203-7214. 10.3748/wjg.v22.i32.7203.
Dekker, E., Tanis, P.J., Vleugels, J.L.A., Kasi, P.M., and Wallace, M.B. (2019). Colorectal cancer.
Lancet 394, 1467-1480. 10.1016/s0140-6736(19)32319-0.
Deryugina, E.I., and Kiosses, W.B. (2017). Intratumoral Cancer Cell Intravasation can occur
Independent of Invasion into the Adjacent Stroma. Cell Rep 19, 601-616.
10.1016/j.celrep.2017.03.064.
Driehuis, E., van Hoeck, A., Moore, K., Kolders, S., Francies, H.E., Gulersonmez, M.C., Stigter,
E.C.A., Burgering, B., Geurts, V., Gracanin, A., et al. (2019). Pancreatic cancer organoids
recapitulate disease and allow personalized drug screening. Proc Natl Acad Sci U S A.
10.1073/pnas.1911273116.
Drost, J., and Clevers, H. (2018). Organoids in cancer research. Nat Rev Cancer 18, 407-418.
10.1038/s41568-018-0007-6.
Farshidfar, F., Weljie, A.M., Kopciuk, K.A., Hilsden, R., McGregor, S.E., Buie, W.D., MacLean, A.,
Vogel, H.J., and Bathe, O.F. (2016). A validated metabolomic signature for colorectal cancer:
exploration of the clinical value of metabolomics. Br J Cancer 115, 848-857.
10.1038/bjc.2016.243.
Faubert, B., Solmonson, A., and DeBerardinis, R.J. (2020). Metabolic reprogramming and cancer
progression. Science 368. 10.1126/science.aaw5473.
Follain, G., Herrmann, D., Harlepp, S., Hyenne, V., Osmani, N., Warren, S.C., Timpson, P., and
Goetz, J.G. (2020). Fluids and their mechanics in tumour transit: shaping metastasis. Nat Rev
Cancer 20, 107-124. 10.1038/s41568-019-0221-x.
Follain, G., Osmani, N., Azevedo, A.S., Allio, G., Mercier, L., Karreman, M.A., Solecki, G., Garcia
Leon, M.J., Lefebvre, O., Fekonja, N., et al. (2018). Hemodynamic Forces Tune the Arrest,
Adhesion, and Extravasation of Circulating Tumor Cells. Dev Cell 45, 33-52.e12.
10.1016/j.devcel.2018.02.015.
Gaggioli, C., Hooper, S., Hidalgo-Carcedo, C., Grosse, R., Marshall, J.F., Harrington, K., and Sahai,
E. (2007). Fibroblast-led collective invasion of carcinoma cells with differing roles for
RhoGTPases in leading and following cells. Nat Cell Biol 9, 1392-1400. 10.1038/ncb1658.
Gayer, C.P., and Basson, M.D. (2009). The effects of mechanical forces on intestinal physiology
and pathology. Cell Signal 21, 1237-1244. 10.1016/j.cellsig.2009.02.011.
Gould, S.E., Junttila, M.R., and de Sauvage, F.J. (2015). Translational value of mouse models in
oncology drug development. Nat Med 21, 431-439. 10.1038/nm.3853.
Grassart, A., Malarde, V., Gobaa, S., Sartori-Rupp, A., Kerns, J., Karalis, K., Marteyn, B.,
Sansonetti, P., and Sauvonnet, N. (2019). Bioengineered Human Organ-on-Chip Reveals
Intestinal Microenvironment and Mechanical Forces Impacting Shigella Infection. Cell Host
Microbe 26, 435-444.e434. 10.1016/j.chom.2019.08.007.
102
Hachey, S.J., Movsesyan, S., Nguyen, Q.H., Burton-Sojo, G., Tankazyan, A., Wu, J., Hoang, T.,
Zhao, D., Wang, S., Hatch, M.M., et al. (2021). An in vitro vascularized micro-tumor model of
human colorectal cancer recapitulates in vivo responses to standard-of-care therapy. Lab Chip.
10.1039/d0lc01216e.
Hapach, L.A., Mosier, J.A., Wang, W., and Reinhart-King, C.A. (2019). Engineered models to
parse apart the metastatic cascade. NPJ Precis Oncol 3, 20. 10.1038/s41698-019-0092-3.
Hassell, B.A., Goyal, G., Lee, E., Sontheimer-Phelps, A., Levy, O., Chen, C.S., and Ingber, D.E.
(2017). Human Organ Chip Models Recapitulate Orthotopic Lung Cancer Growth, Therapeutic
Responses, and Tumor Dormancy In Vitro. Cell Rep 21, 508-516. 10.1016/j.celrep.2017.09.043.
Hatzikirou, H., Basanta, D., Simon, M., Schaller, K., and Deutsch, A. (2012). 'Go or grow': the key
to the emergence of invasion in tumour progression? Math Med Biol 29, 49-65.
10.1093/imammb/dqq011.
Herrera, M., Islam, A.B., Herrera, A., Martín, P., García, V., Silva, J., Garcia, J.M., Salas, C., Casal,
I., de Herreros, A.G., et al. (2013). Functional heterogeneity of cancer-associated fibroblasts
from human colon tumors shows specific prognostic gene expression signature. Clin Cancer Res
19, 5914-5926. 10.1158/1078-0432.CCR-13-0694.
Huang, J.W., Pan, H.J., Yao, W.Y., Tsao, Y.W., Liao, W.Y., Wu, C.W., Tung, Y.C., and Lee, C.H.
(2013). Interaction between lung cancer cell and myofibroblast influenced by cyclic tensile
strain. Lab Chip 13, 1114-1120. 10.1039/c2lc41050h.
Huh, D., Kim, H.J., Fraser, J.P., Shea, D.E., Khan, M., Bahinski, A., Hamilton, G.A., and Ingber, D.E.
(2013). Microfabrication of human organs-on-chips. Nat Protoc 8, 2135-2157.
10.1038/nprot.2013.137.
Humphries, A., and Wright, N.A. (2008). Colonic crypt organization and tumorigenesis. Nat Rev
Cancer 8, 415-424. 10.1038/nrc2392.
Hyland, N.P., and Cryan, J.F. (2010). A Gut Feeling about GABA: Focus on GABA(B) Receptors.
Front Pharmacol 1, 124. 10.3389/fphar.2010.00124.
Jalili-Firoozinezhad, S., Prantil-Baun, R., Jiang, A., Potla, R., Mammoto, T., Weaver, J.C.,
Ferrante, T.C., Kim, H.J., Cabral, J.M.S., Levy, O., and Ingber, D.E. (2018). Modeling radiation
injury-induced cell death and countermeasure drug responses in a human Gut-on-a-Chip. Cell
Death Dis 9, 223. 10.1038/s41419-018-0304-8.
Jang, K.J., Otieno, M.A., Ronxhi, J., Lim, H.K., Ewart, L., Kodella, K.R., Petropolis, D.B., Kulkarni,
G., Rubins, J.E., Conegliano, D., et al. (2019). Reproducing human and cross-species drug
toxicities using a Liver-Chip. Sci Transl Med 11, 1-12. 10.1126/scitranslmed.aax5516.
Jiang, S.H., Hu, L.P., Wang, X., Li, J., and Zhang, Z.G. (2020). Neurotransmitters: emerging targets
in cancer. Oncogene 39, 503-515. 10.1038/s41388-019-1006-0.
Karnoub, A.E., Dash, A.B., Vo, A.P., Sullivan, A., Brooks, M.W., Bell, G.W., Richardson, A.L.,
Polyak, K., Tubo, R., and Weinberg, R.A. (2007). Mesenchymal stem cells within tumour stroma
promote breast cancer metastasis. Nature 449, 557-563. 10.1038/nature06188.
Kasendra, M., Luc, R., Yin, J., Manatakis, D.V., Kulkarni, G., Lucchesi, C., Sliz, J., Apostolou, A.,
Sunuwar, L., Obrigewitch, J., et al. (2020). Duodenum Intestine-Chip for preclinical drug
assessment in a human relevant model. Elife 9. 10.7554/eLife.50135.
Kasendra, M., Tovaglieri, A., Sontheimer-Phelps, A., Jalili-Firoozinezhad, S., Bein, A., Chalkiadaki,
A., Scholl, W., Zhang, C., Rickner, H., Richmond, C.A., et al. (2018). Development of a primary
103
human Small Intestine-on-a-Chip using biopsy-derived organoids. Sci Rep 8, 2871.
10.1038/s41598-018-21201-7.
Kim, D., Paggi, J.M., Park, C., Bennett, C., and Salzberg, S.L. (2019). Graph-based genome
alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol 37, 907-915.
10.1038/s41587-019-0201-4.
Kim, H.J., Huh, D., Hamilton, G., and Ingber, D.E. (2012). Human gut-on-a-chip inhabited by
microbial flora that experiences intestinal peristalsis-like motions and flow. Lab Chip 12, 2165-
2174. 10.1039/c2lc40074j.
Kim, H.J., Li, H., Collins, J.J., and Ingber, D.E. (2016). Contributions of microbiome and
mechanical deformation to intestinal bacterial overgrowth and inflammation in a human gut-
on-a-chip. Proc Natl Acad Sci U S A 113, E7-15. 10.1073/pnas.1522193112.
Kleinrok, Z., Matuszek, M., Jesipowicz, J., Matuszek, B., Opolski, A., and Radzikowski, C. (1998).
GABA content and GAD activity in colon tumors taken from patients with colon cancer or from
xenografted human colon cancer cells growing as s.c. tumors in athymic nu/nu mice. J Physiol
Pharmacol 49, 303-310.
Ledford, H. (2011). Translational research: 4 ways to fix the clinical trial. In Nature, pp. 526-528.
10.1038/477526a.
Lee, S.W.L., Adriani, G., Ceccarello, E., Pavesi, A., Tan, A.T., Bertoletti, A., Kamm, R.D., and
Wong, S.C. (2018). Characterizing the Role of Monocytes in T Cell Cancer Immunotherapy Using
a 3D Microfluidic Model. Front Immunol 9, 416. 10.3389/fimmu.2018.00416.
Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and
dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550. 10.1186/s13059-014-0550-8.
Lüddens, H., Pritchett, D.B., Köhler, M., Killisch, I., Keinänen, K., Monyer, H., Sprengel, R., and
Seeburg, P.H. (1990). Cerebellar GABAA receptor selective for a behavioural alcohol antagonist.
Nature 346, 648-651. 10.1038/346648a0.
Mak, I.W., Evaniew, N., and Ghert, M. (2014). Lost in translation: animal models and clinical
trials in cancer treatment. Am J Transl Res 6, 114-118.
McAleer, C.W., Long, C.J., Elbrecht, D., Sasserath, T., Bridges, L.R., Rumsey, J.W., Martin, C.,
Schnepper, M., Wang, Y., Schuler, F., et al. (2019). Multi-organ system for the evaluation of
efficacy and off-target toxicity of anticancer therapeutics. Sci Transl Med 11.
10.1126/scitranslmed.aav1386.
Nguyen, D.T., Lee, E., Alimperti, S., Norgard, R.J., Wong, A., Lee, J.J., Eyckmans, J., Stanger, B.Z.,
and Chen, C.S. (2019). A biomimetic pancreatic cancer on-chip reveals endothelial ablation via
ALK7 signaling. Sci Adv 5, eaav6789. 10.1126/sciadv.aav6789.
Nie, X., Liu, H., Liu, L., Wang, Y.D., and Chen, W.D. (2020). Emerging Roles of Wnt Ligands in
Human Colorectal Cancer. Front Oncol 10, 1341. 10.3389/fonc.2020.01341.
Niu, G., Deng, L., Zhang, X., Hu, Z., Han, S., Xu, K., Hong, R., Meng, H., and Ke, C. (2020). GABRD
promotes progression and predicts poor prognosis in colorectal cancer. Open Med (Wars) 15,
1172-1183. 10.1515/med-2020-0128.
Ohlund, D., Handly-Santana, A., Biffi, G., Elyada, E., Almeida, A.S., Ponz-Sarvise, M., Corbo, V.,
Oni, T.E., Hearn, S.A., Lee, E.J., et al. (2017). Distinct populations of inflammatory fibroblasts
and myofibroblasts in pancreatic cancer. J Exp Med 214, 579-596. 10.1084/jem.20162024.
Olejniczak, A., Szarynska, M., and Kmiec, Z. (2018). In vitro characterization of spheres derived
from colorectal cancer cell lines. Int J Oncol 52, 599-612. 10.3892/ijo.2017.4206.
104
Ooft, S.N., Weeber, F., Dijkstra, K.K., McLean, C.M., Kaing, S., van Werkhoven, E., Schipper, L.,
Hoes, L., Vis, D.J., van de Haar, J., et al. (2019). Patient-derived organoids can predict response
to chemotherapy in metastatic colorectal cancer patients. Sci Transl Med 11.
10.1126/scitranslmed.aay2574.
Orimo, A., Gupta, P.B., Sgroi, D.C., Arenzana-Seisdedos, F., Delaunay, T., Naeem, R., Carey, V.J.,
Richardson, A.L., and Weinberg, R.A. (2005). Stromal fibroblasts present in invasive human
breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12
secretion. Cell 121, 335-348. 10.1016/j.cell.2005.02.034.
Osmani, N., Follain, G., Garcia Leon, M.J., Lefebvre, O., Busnelli, I., Larnicol, A., Harlepp, S., and
Goetz, J.G. (2019). Metastatic Tumor Cells Exploit Their Adhesion Repertoire to Counteract
Shear Forces during Intravascular Arrest. Cell Rep 28, 2491-2500.e2495.
10.1016/j.celrep.2019.07.102.
Osmulski, P.A., Cunsolo, A., Chen, M., Qian, Y., Lin, C.L., Hung, C.N., Mahalingam, D., Kirma,
N.B., Chen, C.L., Taverna, J.A., et al. (2021). Contacts with Macrophages Promote an Aggressive
Nanomechanical Phenotype of Circulating Tumor Cells in Prostate Cancer. Cancer Res 81, 4110-
4123. 10.1158/0008-5472.CAN-20-3595.
Pavesi, A., Tan, A.T., Koh, S., Chia, A., Colombo, M., Antonecchia, E., Miccolis, C., Ceccarello, E.,
Adriani, G., Raimondi, M.T., et al. (2017). A 3D microfluidic model for preclinical evaluation of
TCR-engineered T cells against solid tumors. JCI Insight 2. 10.1172/jci.insight.89762.
Peterson, M.D., and Mooseker, M.S. (1992). Characterization of the enterocyte-like brush
border cytoskeleton of the C2BBe clones of the human intestinal cell line, Caco-2. J Cell Sci 102 (
Pt 3), 581-600.
Pino, M.S., Kikuchi, H., Zeng, M., Herraiz, M.T., Sperduti, I., Berger, D., Park, D.Y., Iafrate, A.J.,
Zukerberg, L.R., and Chung, D.C. (2010). Epithelial to mesenchymal transition is impaired in
colon cancer cells with microsatellite instability. Gastroenterology 138, 1406-1417.
10.1053/j.gastro.2009.12.010.
Qian, J., Olbrecht, S., Boeckx, B., Vos, H., Laoui, D., Etlioglu, E., Wauters, E., Pomella, V.,
Verbandt, S., Busschaert, P., et al. (2020). A pan-cancer blueprint of the heterogeneous tumor
microenvironment revealed by single-cell profiling. Cell Res 30, 745-762. 10.1038/s41422-020-
0355-0.
Quail, D., and Joyce, J. (2013). Microenvironmental regulation of tumor progression and
metastasis. Nat Med 19, 1423-1437. 10.1038/nm.3394.
Ramzy, G.M., Koessler, T., Ducrey, E., McKee, T., Ris, F., Buchs, N., Rubbia-Brandt, L., Dietrich,
P.Y., and Nowak-Sliwinska, P. (2020). Patient-Derived In Vitro Models for Drug Discovery in
Colorectal Carcinoma. Cancers (Basel) 12. 10.3390/cancers12061423.
Sahai, E., Astsaturov, I., Cukierman, E., DeNardo, D.G., Egeblad, M., Evans, R.M., Fearon, D.,
Greten, F.R., Hingorani, S.R., Hunter, T., et al. (2020). A framework for advancing our
understanding of cancer-associated fibroblasts. Nat Rev Cancer 20, 174-186. 10.1038/s41568-
019-0238-1.
Sarvestani, S.K., DeHaan, R.K., Miller, P.G., Bose, S., Shen, X., Shuler, M.L., and Huang, E.H.
(2020). A Tissue Engineering Approach to Metastatic Colon Cancer. iScience 23, 101719.
10.1016/j.isci.2020.101719.
Sato, T., Stange, D.E., Ferrante, M., Vries, R.G., Van Es, J.H., Van den Brink, S., Van Houdt, W.J.,
Pronk, A., Van Gorp, J., Siersema, P.D., and Clevers, H. (2011). Long-term expansion of epithelial
105
organoids from human colon, adenoma, adenocarcinoma, and Barrett's epithelium.
Gastroenterology 141, 1762-1772. 10.1053/j.gastro.2011.07.050.
Sato, T., Vries, R.G., Snippert, H.J., van de Wetering, M., Barker, N., Stange, D.E., van Es, J.H.,
Abo, A., Kujala, P., Peters, P.J., and Clevers, H. (2009). Single Lgr5 stem cells build crypt-villus
structures in vitro without a mesenchymal niche. Nature 459, 262-265. 10.1038/nature07935.
Shen, Y., Wang, X., Lu, J., Salfenmoser, M., Wirsik, N.M., Schleussner, N., Imle, A., Freire Valls,
A., Radhakrishnan, P., Liang, J., et al. (2020). Reduction of Liver Metastasis Stiffness Improves
Response to Bevacizumab in Metastatic Colorectal Cancer. Cancer Cell 37, 800-817.e807.
10.1016/j.ccell.2020.05.005.
Shetty, S., Lalor, P.F., and Adams, D.H. (2018). Liver sinusoidal endothelial cells - gatekeepers of
hepatic immunity. Nat Rev Gastroenterol Hepatol 15, 555-567. 10.1038/s41575-018-0020-y.
Shu, Q., Liu, J., Liu, X., Zhao, S., Li, H., Tan, Y., and Xu, J. (2016). GABAB R/GSK-3β/NF-κB
signaling pathway regulates the proliferation of colorectal cancer cells. Cancer Med 5, 1259-
1267. 10.1002/cam4.686.
Siegel, R.L., Miller, K.D., Fuchs, H.E., and Jemal, A. (2021). Cancer Statistics, 2021. CA Cancer J
Clin 71, 7-33. 10.3322/caac.21654.
Siegel, R.L., Miller, K.D., Goding Sauer, A., Fedewa, S.A., Butterly, L.F., Anderson, J.C., Cercek, A.,
Smith, R.A., and Jemal, A. (2020). Colorectal cancer statistics, 2020. CA Cancer J Clin.
10.3322/caac.21601.
Skardal, A., Devarasetty, M., Forsythe, S., Atala, A., and Soker, S. (2016). A reductionist
metastasis-on-a-chip platform for in vitro tumor progression modeling and drug screening.
Biotechnol Bioeng 113, 2020-2032. 10.1002/bit.25950.
Sontheimer-Phelps, A., Chou, D.B., Tovaglieri, A., Ferrante, T.C., Duckworth, T., Fadel, C.,
Frismantas, V., Sutherland, A.D., Jalili-Firoozinezhad, S., Kasendra, M., et al. (2020). Human
Colon-on-a-Chip Enables Continuous In Vitro Analysis of Colon Mucus Layer Accumulation and
Physiology. Cell Mol Gastroenterol Hepatol 9, 507-526. 10.1016/j.jcmgh.2019.11.008.
Sontheimer-Phelps, A., Hassell, B.A., and Ingber, D.E. (2019). Modelling cancer in microfluidic
human organs-on-chips. Nat Rev Cancer 19, 65-81. 10.1038/s41568-018-0104-6.
Strelez, C., Chilakala, S., Ghaffarian, K., Lau, R., Spiller, E., Ung, N., Hixon, D., Yoon, A.Y., Sun,
R.X., Lenz, H.J., et al. (2021). Human colorectal cancer-on-chip model to study the
microenvironmental influence on early metastatic spread. iScience 24, 102509.
10.1016/j.isci.2021.102509.
Sun, Y., Li, M., Liu, G., Zhang, X., Zhi, L., Zhao, J., and Wang, G. (2020). The function of Piezo1 in
colon cancer metastasis and its potential regulatory mechanism. J Cancer Res Clin Oncol 146,
1139-1152. 10.1007/s00432-020-03179-w.
Taftaf, R., Liu, X., Singh, S., Jia, Y., Dashzeveg, N.K., Hoffmann, A.D., El-Shennawy, L., Ramos,
E.K., Adorno-Cruz, V., Schuster, E.J., et al. (2021). ICAM1 initiates CTC cluster formation and
trans-endothelial migration in lung metastasis of breast cancer. Nat Commun 12, 4867.
10.1038/s41467-021-25189-z.
Tan, B., Qiu, Y., Zou, X., Chen, T., Xie, G., Cheng, Y., Dong, T., Zhao, L., Feng, B., Hu, X., et al.
(2013). Metabonomics identifies serum metabolite markers of colorectal cancer. J Proteome
Res 12, 3000-3009. 10.1021/pr400337b.
106
Thaker, P.H., Yokoi, K., Jennings, N.B., Li, Y., Rebhun, R.B., Rousseau, D.L., Fan, D., and Sood,
A.K. (2005). Inhibition of experimental colon cancer metastasis by the GABA-receptor agonist
nembutal. Cancer Biol Ther 4, 753-758. 10.4161/cbt.4.7.1827.
Valderrama-Treviño, A.I., Barrera-Mera, B., Ceballos-Villalva, J.C., and Montalvo-Javé, E.E.
(2017). Hepatic Metastasis from Colorectal Cancer. Euroasian J Hepatogastroenterol 7, 166-175.
10.5005/jp-journals-10018-1241.
Vlachogiannis, G., Hedayat, S., Vatsiou, A., Jamin, Y., Fernandez-Mateos, J., Khan, K., Lampis, A.,
Eason, K., Huntingford, I., Burke, R., et al. (2018). Patient-derived organoids model treatment
response of metastatic gastrointestinal cancers. Science 359, 920-926.
10.1126/science.aao2774.
Walrath, J.C., Hawes, J.J., Van Dyke, T., and Reilly, K.M. (2010). Genetically Engineered Mouse
Models in Cancer Research. Adv Cancer Res 106, 113-164. 10.1016/s0065-230x(10)06004-5.
Wu, M., Kim, K.Y., Park, W.C., Ryu, H.S., Choi, S.C., Kim, M.S., Myung, J.Y., Choi, H.S., Kim, E.J.,
and Lee, M.Y. (2020). Enhanced expression of GABRD predicts poor prognosis in patients with
colon adenocarcinoma. Transl Oncol 13, 100861. 10.1016/j.tranon.2020.100861.
Wyckoff, J.B., Jones, J.G., Condeelis, J.S., and Segall, J.E. (2000). A critical step in metastasis: in
vivo analysis of intravasation at the primary tumor. Cancer Res 60, 2504-2511.
Yan, H., Tang, G., Wang, H., Hao, L., He, T., Sun, X., Ting, A.H., Deng, A., and Sun, S. (2016). DNA
methylation reactivates GAD1 expression in cancer by preventing CTCF-mediated polycomb
repressive complex 2 recruitment. Oncogene 35, 3995-4008. 10.1038/onc.2015.423.
Yan, L., Gong, Y.Z., Shao, M.N., Ruan, G.T., Xie, H.L., Liao, X.W., Wang, X.K., Han, Q.F., Zhou, X.,
Zhu, L.C., et al. (2020). Distinct diagnostic and prognostic values of γ-aminobutyric acid type A
receptor family genes in patients with colon adenocarcinoma. Oncol Lett 20, 275-291.
10.3892/ol.2020.11573.
Yao, Y., Xu, X., Yang, L., Zhu, J., Wan, J., Shen, L., Xia, F., Fu, G., Deng, Y., Pan, M., et al. (2019).
Patient-Derived Organoids Predict Chemoradiation Responses of Locally Advanced Rectal
Cancer. Cell Stem Cell. 10.1016/j.stem.2019.10.010.
Yazulla, S., and Brecha, N. (1981). Localized binding of [3H]muscimol to synapses in chicken
retina. Proc Natl Acad Sci U S A 78, 643-647. 10.1073/pnas.78.1.643.
Ying, L., Zhu, Z., Xu, Z., He, T., Li, E., Guo, Z., Liu, F., Jiang, C., and Wang, Q. (2015). Cancer
Associated Fibroblast-Derived Hepatocyte Growth Factor Inhibits the Paclitaxel-Induced
Apoptosis of Lung Cancer A549 Cells by Up-Regulating the PI3K/Akt and GRP78 Signaling on a
Microfluidic Platform. PLoS One 10, e0129593. 10.1371/journal.pone.0129593.
Yu, G., Wang, L.G., Han, Y., and He, Q.Y. (2012). clusterProfiler: an R package for comparing
biological themes among gene clusters. OMICS 16, 284-287. 10.1089/omi.2011.0118.
Zhang, X., and Jonassen, I. (2020). RASflow: an RNA-Seq analysis workflow with Snakemake.
BMC Bioinformatics 21, 110. 10.1186/s12859-020-3433-x.
Abstract (if available)
Abstract
Colorectal cancer (CRC) is one of the deadliest cancers in the U.S., yet we still understand very little about the mechanisms behind this disease. Therefore, I am developing a CRC-Chip model that recapitulates the complex nature of progression to increase our understanding of CRC and accelerate the discovery of new treatments. The organ-on-chip technology developed by Emulate, Inc. maintains physiologically relevant aspects of organ structure and function by incorporating tissue compartments and mechanical forces to mimic in vivo peristalsis and fluid flow. The CRC-Chip consists of two microfluidic compartments separated by a porous membrane, with endothelial cells in the bottom channel and normal colon epithelial cells plus fluorescently-labeled CRC cell lines in the top channel. I optimized various iterations of the CRC-Chip to include CRC cell lines, the addition of patient-derived cancer associated fibroblasts (CAFs), and patient-derived CRC organoids. Several assays were developed to monitor cancer progression: an imaging-based invasion assay in which cancer cells are visualized traversing from the epithelial compartment into the blood vessel compartment via confocal microscopy; and a mass spectrometry-based metabolomics analysis of the effluent to monitor changes in metabolites over time. In this dissertation I show that 1) the CRC-Chip recapitulates the varying aggressiveness of colon cancers and cancer biology, 2) the CRC-Chip allows for studying cancer associated fibroblast (CAF) functional heterogeneity and 3) peristalsis-like mechanical forces influences the invasive capabilities of tumor cells through GABAergic signaling, something that could not previously be studied. The work described here with a unique model system sets the stage for identifying new therapies targeting the stroma and cancer cells responding to a mechanical microenvironment.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Investigating the complexity of the tumor microenvironment's role in drug response
PDF
Exploration of the roles of cancer stem cells and survivin in the pathogenesis and progression of prostate cancer
PDF
Study of bone morphogenetic protein-2 and stromal cell derived factor-1 in prostate cancer
PDF
pRB structures and functions mediating suppression of retinoblastoma initiation
PDF
Microenvironmental and biomechanical regulation of mitochondrial membrane potential in cancer cells
PDF
Genome-scale modeling of macrophage activity in the colorectal cancer microenvironment
PDF
Molecular role of EZH2 overexpression in Colorectal Cancer progression
PDF
PTEN deletion induced tumor initiating cells: Strategies to accelerate the disease progression of liver cancer
PDF
Contribution of cancer associated fibroblasts to cancer progression
PDF
Bimodal effects of bone morphogenetic proteins in prostate cancer
PDF
RNA methylation in cancer plasticity and drug resistance
PDF
A mutant recombinant immunotoxin αFAP-PE38 for melanoma treatment in mice
PDF
Ectopic expression of a truncated isoform of hair keratin 81 in breast cancer alters biophysical characteristics to promote metastatic propensity
PDF
Functional analysis of a prostate cancer risk enhancer at 7p15.2
PDF
Characterization of a new chromobox protein 8 (CBX8) antagonist in a model of human colon cancer
PDF
Engineering immunotoxin and viral vectors for cancer therapy
PDF
Functional characterization of colon cancer risk-associated enhancers: connecting risk loci to risk genes
PDF
Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells
PDF
Functional characterization of colon cancer risk enhancers
PDF
Cisplatin activates mitochondrial oxphos leading to acute treatment resistance in bladder cancer
Asset Metadata
Creator
Strelez, Carly
(author)
Core Title
Development of a colorectal cancer-on-chip to investigate the tumor microenvironment's role in cancer progression
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Cancer Biology and Genomics
Degree Conferral Date
2022-05
Publication Date
01/14/2023
Defense Date
11/09/2021
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
biomimetic model,cancer,cancer associated fibroblast,OAI-PMH Harvest,organ-on-chip,tumor microenvironment
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Farnham , Peggy (
committee chair
), Frey, Mark (
committee member
), Mumenthaler, Shannon (
committee member
), Stiles, Bangyan (
committee member
)
Creator Email
carlystrelez@gmail.com,strelez@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC110520285
Unique identifier
UC110520285
Legacy Identifier
etd-StrelezCar-10344
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Strelez, Carly
Type
texts
Source
20220124-usctheses-batch-908
(batch),
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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.
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
Repository Email
cisadmin@lib.usc.edu
Tags
biomimetic model
cancer associated fibroblast
organ-on-chip
tumor microenvironment