ENGINEERED ORGANOTYPIC BREAST TUMOR MODEL FOR MECHANISTIC
STUDIES OF TUMOR-STROMAL INTERACTIONS AND DRUG DISCOVERY
A Dissertation
Presented to
The Graduate Faculty of The University of Akron
In Partial Fulfillment
of the Requirements for the Degree
Doctor of Philosophy
Sunil Singh
May, 2021
ENGINEERED ORGANOTYPIC BREAST TUMOR MODEL FOR MECHANISTIC
STUDIES OF TUMOR-STROMAL INTERACTIONS AND DRUG DISCOVERY
Sunil Singh
Dissertation
Approved: Accepted:
______Advisor Interim Department Chair Dr. Hossein Tavana Dr. Francis Loth
______Committee Member Interim Dean of the College Dr. Marnie M. Saunders Dr. Craig Menzemer
______Committee Member Interim Director, Graduate School Dr. Nic D. Leipzig Dr. Marnie Saunders
______Committee Member Date Dr. Francis Loth
______Committee Member Dr. Adam W. Smith ii
ABSTRACT
Cancer is the second leading cause of mortality in the United States. The
National Cancer Institute estimated 1.7 million new cases of cancer and 0.6 million cancer deaths in the United States in 2019. Cancer is a heterogeneous disease that involves not only cancer cells, but also different cells and proteins in the tumor stroma. Various cells such as fibroblasts, infiltrating immune cells, and endothelial cells, the extracellular matrix (ECM) proteins, growth factors, chemokines, cytokines, and other bioactive agents constitute the tumor microenvironment
(TME). Traditional cancer treatments only target cancer cells but growing evidence has established pivotal roles for the TME in driving tumor progression and chemoresistance. As such, targeting the TME and its interactions with cancer cells
(known as tumor-stromal interactions) is now being pursued as a new approach to improve treatment outcomes for patients. However, preclinical models such as standard in vitro cell cultures and animal models routinely used in cancer drug discovery fail to recapitulate tumor-stromal interactions of human tumors.
To overcome limitations of existing tumor models, we developed a three- dimensional (3D) organotypic tumor model that enables mechanistic studies of tumor-stromal interactions to enable drug discovery efforts against the TME. The organotypic model incorporates three key components of the TME: a mass of
iii cancer cells, fibroblasts, and collagen as the ECM. The resulting model resembles the architecture of solid tumors and spatial distribution of cells within the TME. We used a novel cell and protein micropatterning approach based on an aqueous two- phase system (ATPS) to first generate a cancer cell spheroid and then overlay it with a collagen solution containing fibroblasts. We automated this technology and adapted it to a high throughput 384-well plate format to enable both mechanistic and phenotypic studies of tumor-stromal interactions and testing arrays of drugs.
We focused on triple negative breast cancer (TNBC) as a disease model because TNBC is the most aggressive subtype of breast cancer with very limited targeted therapy options and poor patient outcomes from cytotoxic chemotherapies, underscoring an unmet need for new treatment strategies. We leveraged our organotypic tumor model to demonstrate the feasibility of mechanistic studies of tumor-stromal interactions in TNBC, focusing on cancer- associated fibroblasts (CAFs) as the most abundant stromal cells in breast tumors.
To establish the validity of our model, we used a well-known chemokine-receptor interaction mechanism in TNBC. Specifically, we showed that fibroblasts-secreted
CXCL12 chemokine promotes the ECM invasion of CXCR4+ TNBC cells by activating oncogenic mitogen-activated protein kinase (MAPK) pathway.
Additionally, the fibroblast cells remodeled the ECM through the
RhoA/ROCK/myosin light chain-2 pathway. Following the validation step, we incorporated patient-derived CAFs in our model and studied their dynamic interactions with TNBC cells to explore whether CAFs-TNBC interactions would
iv present therapy targets. Our mechanistic studies showed that hepatocyte growth factor (HGF) secreted by CAFs predominantly activates MET receptor tyrosine kinase on TNBC cells to promote proliferation, invasiveness, and epithelial-to- mesenchymal transition (EMT) of TNBC cells. This interaction axis led to activation of oncogenic pathways such as MAPK, phosphatidylinositol 3-kinase-Akt
(PI3K/Akt), and signal transducer and activator of transcription (STAT) in TNBC cells. Importantly, we found that TNBC cells become resistant to single-agent treatment with a potent MAPK pathway inhibitor (trametinib) and demonstrated a design-driven approach to select drug combinations that effectively inhibit pro- metastatic functions of TNBC cells. We also demonstrated that the HGF-MET axis is implicated in lung metastasis of TNBC and that blocking this signaling axis is a potential approach against both primary TNBC tumorigenesis and metastases formation in the lung. Future studies are needed to study long-term effectiveness of drug combinations in our organotypic tumor model.
Overall, this work established the utility of our 3D organotypic tumor model to elucidate the role of tumor stroma in promoting pro-metastatic functions of TNBC cells, thereby facilitating the design and development of novel therapeutic approaches against tumor-stromal interactions. This technology will facilitate future studies to incorporate other components of tumor stroma and patient- derived cancer and stromal cells to expedite the translation of the findings.
v
ACKNOWLEDGEMENT
First and foremost, I would like to express my sincere gratitude to my advisor, Dr. Hossein Tavana, for his enthusiastic support and continuous guidance throughout my Ph.D. program at The University of Akron. Without his persistent help and dedicated involvement in my research projects, this dissertation would not have been possible. I could not have imagined having a better advisor and mentor for my Ph.D. study.
I would also like to thank Dr. Marnie M. Saunders, Dr. Nic D. Leipzig, Dr.
Francis Loth, and Dr. Adam W. Smith for serving as my committee members and for their support and feedback to improve the quality of my research. My sincere thanks to my lab members for creating a positive, helpful, and productive environment.
I dedicate my dissertation work to my family. A special feeling of gratitude to my parents, Usha and Babu Raja Singh, who have loved me unconditionally and taught me to dream big. They have been instrumental in my decision to pursue higher education. This work is also dedicated to my beautiful wife, Arsree
Shrestha, who has been by my side through thick and thin. You are an embodiment of love and kindness. You are my rock.
vi
TABLE OF CONTENTS
LIST OF FIGURES ...... xi
LIST OF TABLES ...... xxv
CHAPTER
I. INTRODUCTION ...... 1
1.1 Significance ...... 1
1.2 Triple Negative Breast Cancer (TNBC) ...... 3
1.3 Tumor Microenvironment (TME) ...... 4
1.3.1 Components of TME ...... 5
1.3.2 Tumor Stroma as Therapeutic Target ...... 9
1.4 Models of Tumor-Stromal Interactions ...... 11
1.4.1 2D Cell Culture Models ...... 11
1.4.2 Animal Models ...... 12
1.4.3 3D Cell Culture Models ...... 13
1.5 Aims and Scope ...... 19
II. COLLAGEN PARTITION IN POLYMERIC AQUEOUS TWO-PHASE SYSTEMS FOR TISSUE ENGINEERINGS ...... 24 2.1 Materials and Methods ...... 27
2.1.1 Preparation of Aqueous Two-Phase Systems (ATPS) ...... 27
2.1.2 Construction of Binodal Curves...... 27
vii 2.1.3 Cell Culture ...... 28
2.1.4 Preparation of Collagen Solution ...... 28
2.1.5 Partition of Collagen in ATPS ...... 29
2.1.6 Hydroxyproline Assay ...... 29
2.1.7 Cancer Cell Spheroid Formation ...... 30
2.1.8 Formation of Collagen-Embedded Cancer Cell Spheroids ...... 31
2.1.9 Statistical Analysis ...... 32
2.2 Results ...... 32
2.2.1 Characterization of ATPS ...... 32
2.2.2 Standard Curves of Collagen Concentration Using Hydroxyproline Assay ...... 33
2.2.3 Collagen Partition in ATPS ...... 34
2.2.4 Collagen Partitioning for Tumor Tissue Engineering ...... 35
2.3 Discussion ...... 36
2.4 Summary ...... 39
III. ORGANOTYPIC BREAST TUMOR MODEL ELUCIDATES DYNAMIC REMODELING OF TUMOR MICROENVIRONMENT ...... 47 3.1 Materials and Methods ...... 51
3.1.1 Cell Culture ...... 51
3.1.2 Culture of Fibroblasts in Collagen ...... 52
3.1.3 Spheroid Formation ...... 53
3.1.4 Immunofluorescence of 3D Cultures of Dispersed Fibroblasts ...... 54
3.1.5 Metabolic Activity of Fibroblast Cells in 3D Cultures ...... 55
3.1.6 Conditioned Medium Cultures ...... 55
3.1.7 Wound Healing Assay ...... 55
viii 3.1.8 Organotypic Microtissue Formation, ECM Invasion, and AMD3100 Treatment ...... 56
3.1.9 Western Blotting...... 58
3.1.10 Statistical Analysis ...... 59
3.2 Results ...... 60
3.2.1 Fibroblasts Contract the ECM ...... 60
3.2.2 Mechanism of ECM Contraction by Fibroblasts ...... 62
3.2.3 The Role of Fibroblasts in Matrix Invasion of Cancer Cells ...... 63
3.2.4 Inhibition of ECM Invasion of CXCR4+TNBC ...... 65
3.3 Discussion ...... 66
3.4 Summary ...... 71
IV. ORGANOTYPIC BREAST TUMOR MODEL ENABLES MECHANISTIC STUDIES OF TUMOR-STROMAL INTERACTIONS IN TNBC ...... 88 4.1 Materials and Methods ...... 92
4.1.1 Cell Culture ...... 92
4.1.2 Human phospho-RTK array ...... 93
4.1.3 Bead-Based Multiplex Immunoassay ...... 94
4.1.4 Organotypic Tumor Model Formation ...... 95
4.1.5 Confocal Microscopy of TNBC Cell Invasion in Organotypic Tumor Models ...... 96
4.1.6 Flow Cytometry of Organotypic Cultures ...... 97
4.1.7 Quantitative PCR ...... 97
4.1.8 Western Blotting...... 98
4.1.9 Drug Treatments ...... 99
4.1.10 Cyclic Treatment of TNBC Spheroids ...... 100
4.1.11 Combination Treatments of Organotypic Cultures ...... 101 ix 4.1.12 Colony Formation Assay and Immunofluorescence Imaging of Proliferative TNBC Cells ...... 102
4.1.13 Bioinformatics Analysis of Data from Breast Cancer Patients ...... 103
4.1.14 Gene Set Enrichment Analysis (GSEA) ...... 103
4.1.15 Statistical Analysis ...... 104
4.2. Results ...... 104
4.2.1 CAFs Interact with TNBC Cells via HGF-MET Pathway ...... 104
4.2.2 CAFs Activate Multiple Downstream Pathways in TNBC Cell Lines . 105
4.2.3 MET Expression and Activity in Patient Tumors ...... 106
4.2.4 CAFs Promote Pro-Metastatic Functions in TNBC Cells ...... 107
4.2.5 Modeling Resistance of TNBC Cells to Single-Agent Treatments .... 109
4.2.6 Treatment Strategies to Suppress Pro-Metastatic Functions of TNBC Cells ...... 110
4.2.7 The Role of HGF-MET Pathway in Lung Metastatic Environments... 113
4.3 Discussion ...... 115
4.4 Summary ...... 119
V. CONCLUSIONS ...... 133
VI. FUTURE WORK ...... 136
BIBLIOGRAPHY ...... 141
APPENDIX ...... 174
x LIST OF FIGURES
Figure Page
1.1: Cancer cells secrete soluble factors to activate fibroblasts into cancer-
associated fibroblasts (CAFs), which in turn remodel the tumor
microenvironment and promote tumor growth, invasion, and metastasis.
...... 22
1.2: Phase diagram of ATPS with 35 kDa PEG and 500 kDa DEX. The solid line
represents binodal curve obtained by fitting experimental data (open
circles). The dashed line represents tie-line to show the locations of initial
and final compositions of an ATPS. Combinations of concentrations of PEG
and DEX above the binodal curve will result in two phases. The inset
cartoon shows equilibrated ATPS separated into two distinct PEG-rich and
DEX-rich phases (Adapted with permission from E. Atefi, J. A. Mann, and
H. Tavana, “Ultralow interfacial tensions of aqueous two-phase systems
measured using drop shape,” Langmuir, vol. 30, no. 32, pp. 9691–9699,
2014. Copyright © 2014 American Chemical Society)...... 23
2.1: Binodal curves of four systems made with different molecular weight PEG
and DEX: (a) PEG 8 kDa-DEX 40 kDa, (b) PEG 8 kDa-DEX 500 kDa, (c)
PEG 35 kDa-DEX 40 kDa, and (d) PEG 35 kDa-DEX 500 kDa. Colored
points show the resulting concentrations of PEG and DEX in the partition
assay; square: 5%PEG-7%DEX, triangle: 6%PEG-8%DEX, and circle:
xi 8%PEG-10.2%DEX. All concentrations of aqueous PEG and DEX solutions
were calculated in %(w/v)...... 40
2.2: Standard curves for (a) hydroxyproline assay and (b) for hydroxyproline
standards prepared in distilled water, a 10%(w/v) DEX 500 kDa solution,
and a 10%(w/v) PEG 35 kDa solution. All samples were hydrolyzed, and
absorbance was measured following the manufacturer's protocol. Statistical
method: two-way ANOVA. NS: statistically non-significant. Error bars are
standard error of the mean. n=5 ...... 41
2.3: Schematic of collagen partition in aqueous two-phase systems ...... 42
2.4: Collagen partition assay containing equal volumes of 15% PEG 35 kDa,
21% DEX 500 kDa, and 2 mg/ml collagen resulting in formation of 5% PEG-
7% DEX ATPS. All concentrations of aqueous PEG and DEX solutions were
calculated in %(w/v)...... 43
2.5: Collagen partition coefficient (Kc) in four different systems using the 5%
PEG-7% DEX concentration pair. *p<0.05, n=4. All concentrations of
aqueous PEG and DEX solutions were calculated in %(w/v)...... 44
2.6: Collagen partition coefficient (Kc) in four different systems in 5% PEG-7%
DEX, 6% PEG-8% DEX, and 8% PEG-10.2% DEX concentration pairs.
*p<0.05, n=4. All concentrations of aqueous PEG and DEX solutions were
calculated in %(w/v)...... 45
xii 2.7: (a) Schematic of embedding a spheroid in partitioned collagen. (b) From a
14% DEX 500 kDa solution mixed with 5 µl of 4 mg/ml collagen solution, a
10 µl volume is dispensed in a tube containing 30 µl of 6.6% PEG 35 kDa.
Collagen gels in the bottom phase. (c,d) Top view of a BT474 cancer cell
spheroid embedded in a collagen gel partitioned to the DEX phase in a
microwell. (d) Side view of a BT474 spheroid in a collagen gel partitioned to
the DEX phase in a PCR tube. All concentrations of aqueous PEG and DEX
solutions were calculated in %(w/v)...... 46
3.1: Formation of organotypic culture. (a) Schematic process of formation of a
microtissue containing a CXCR4+TNBC spheroid and fibroblasts dispersed
in collagen. (b) 3D confocal reconstruction of a microtissue containing
CXCR4+TNBC spheroids (blue) and fibroblasts (green). Collagen was not
stained. The numbers in the red box represent distances in micrometers.
...... 73
3.2: Mechanical properties of ECM. (a) Elastic moduli of type I rat tail collagen
gels prepared from different protein concentrations maintained in PBS and
measured using AFM within 4 h of gel formation. (b) A typical histogram
obtained from force curve fitting result for a 4 mg/ml collagen gel. Error bars
are standard error of the mean and n=5...... 74
3.3: Characterize ECM contraction by fibroblasts. (a) 3D view of spatial
distribution of fibroblasts (green) in collagen. (b) Fibroblasts were dispersed
in collagen at a density of 2×103, 4×103, 6×103, or 8×103 cells per well. Each xiii well contained 20 µl of collagen solution. The number of cells in each well
was estimated by analyzing each stack of confocal images in MATLAB.
Briefly, images from each stack were opened in ImageJ and XY coordinates
of the cells were determined in an excel sheet. The coordinates were
imported in MATLAB to count the total number of cells. The cells that had
similar XY coordinates in four consecutive images were considered
redundant and hence counted only once. Each cell XY coordinates usually
repeated in two or three consecutive images. (c, d) Phase contrast images
show contraction of collagen gels, 1 mg/ml and 4 mg/ml, by fibroblasts at
different densities of 5×103 (5k), 10×103 (10k), and 1.5×104 (15k) on day 4
and day 10. Each density indicates the number of cells per well of 384-well
plates. (e, f) Time-dependent contractility of collagen gel, 1 mg/ml and 4
mg/ml, by fibroblasts of different densities (5k,10k, and 15k). (g)
Comparison of matrix contractility of 4 mg/ml collagen gels by 15k density
of HMFs and CXCL12+HMFs. *p<0.05, n=16. Two-tailed unpaired t-test was
used to compare the two experimental groups...... 75
3.4: Mechanism of collagen contraction by fibroblasts. (a) Immunofluorescence
staining of β-actin (green) and Hoechst (blue) in HMFs and CXCL12+HMFs
dispersed in collagen on day 5 of culture. (b) Box plot of cell aspect ratio
( 푊𝑖푑푡ℎ ) for HMFs and CXCL12+HMFs dispersed in collagen. The boxes 퐻푒𝑖𝑔ℎ푡
represent the 25th and 75th percentiles with the median shown with a
horizontal line inside each box. The mean is shown by a cross symbol inside
xiv each box. The whiskers represent the 10th and 90th percentiles of the data.
Two-tailed Mann-Whitney test was used to calculate a p-value. *p<0.01,
n=10. (c) Western blot analysis of expression levels of phosphorylated and
total MLC2 in HMFs and CXCL12+HMFs dispersed in collagen. (d) MLC2
activity quantified as p-MLC2/MLC2 from three separate experiments. Data
were normally distributed, and two-tailed, unpaired t-test was used to
calculate a p-value. *p<0.01...... 76
3.5: Characterization of fibroblasts motility (a) Wound healing assay to study
migration of HMFs and CXCL12+HMFs. (b) Migration of cells is quantified
as gap closure= 1-A2/A1, where A1 is the initial gap area and A2 is the final
gap area. *p<0.05 was calculated using two-tailed, unpaired t-test with
n=12. (c) Western blot analysis of expression levels of contractility-
associated proteins in HMFs and CXCL12+HMFs cultured in collagen with
quantified protein expression after normalizing with GAPDH. Data represent
three separate experiments. *p<0.05 was calculated using two-tailed,
unpaired t-test. (d) Phase contrast images of cultures of fibroblasts
dispersed in collagen gels on day 5 showing inhibition of collagen
contraction using 10 µM of a ROCK inhibitor (Y27632, Selleckchem) ..... 77
3.6: Metabolic activities of HMFs and CXCL12+HMFs dispersed in collagen gels
measured using a PrestoBlue assay. *p<0.05 was calculated using two-
tailed, unpaired t-test with n=16...... 78
xv 3.7: Effect of secretome on fibroblast contractility. (a) Collagen gel containing
dispersed HMFs supplemented with conditioned medium from monoculture
of CXCL12+HMFs. (b) Matrix contractility of HMFs-containing collagen gels
with and without CXCL12+HMFs conditioned medium. *p<0.05 was
calculated using two-tailed, unpaired t-test with n=16...... 79
3.8: Influence of fibroblasts on ECM invasion of cancer cell. (a) Confocal images
of CXCR4+TNBC cell mass in three different microtissues over a 5-day
culture. (b) Matrix contractility for CXCR4+TNBC-containing microtissues
with and without 1.5×104 fibroblasts. Unpaired t-test was used to calculate
the p-values. *p<0.05, **p<0.001, ***p<0.0001, n=10. (c) Normalized
invasion of CXCR4+TNBC cells in microtissues with and without fibroblasts.
Invasion area values were normally distributed and two-tailed, unpaired t-
test was used to calculate the p-values. *p<0.05, **p<0.001, ***p<0.0001,
n=8 ...... 80
3.9: Analysis of ECM invasion of cancer cells along the z-axis. (a) Histograms
of invasion of CXCR4+TNBC cells in three different microtissues on day 3
and day 5. Each colored curve in a graph represents pixels area of
CXCR4+TNBC cells from one microtissue sample and each data point is
obtained from one image of the confocal stack of images of that microtissue.
Histograms have non-Gaussian distribution. (b) Normalized 3D spreading
in three different tumor models. Normalized invasion area and AUC values
xvi were normally distributed. *p<0.05, **p<0.001, and ***p<0.0001 were
calculated using two-tailed, unpaired t-test with n=8...... 81
3.10: XZ images showing the lower hemisphere of a CXCR4+TNBC spheroid
(green) in microtissues containing dispersed (a) HMFs or (b)
CXCL12+HMFs reconstructed from confocal z-stack images ...... 82
3.11: Molecular analysis of TNBC invasiveness. (a) Western blot analysis of
ERK1/2 phosphorylation in different microtissues. (b) Quantified p-ERK1/2
normalized with t-ERK1/2 in five different microtissues. Data represent
three separate experiments. +HMFs and +CXCL12+HMFs indicate
organotypic cultures containing a TNBC spheroid and fibroblasts dispersed
in collagen. For statistics, t-test was used to compare +HMFs and
+CXCL12+HMFs, whereas one-way ANOVA with Tukey’s pairwise
comparisons were used to compare CXCR4+TNBC, HMFs, and
CXCL12+HMFs. *p<0.05 and ***p<0.0001...... 83
3.12: Inhibition of tumor-stromal interactions. (a) Confocal z-projected images of
the CXCR4+TNBC cells in CXCL12+HMFs-containing microtissues without
and with AMD3100 treatment. (b) Normalized invasion area for AMD3100-
treated and non-treated microtissues. *p<0.01 was calculated using two-
tailed, unpaired t-test with n=8. (c) Western blot analysis of effect of
AMD3100 treatment on ERK1/2 activity. The graph shows quantified
phospho-protein levels after normalizing with t-ERK1/2. Data represent
xvii three separate experiments. ***p<0.0001 were calculated using two-tailed,
unpaired t-test ...... 84
3.13: Morphologies of fibroblasts (a) Phase contrast images of HMFs and
CXCL12+HMFs captured 24 h after seeding on collagen gels. (b). Phase
contrast images of HMFs and CXCL12+HMFs dispersed in collagen gels on
day 13 and day 4, respectively, when they have comparable matrix
contractility. HMFs showed a branched morphology, whereas
CXCL12+HMFs had an elongated morphology...... 85
3.14: Parental TNBC cells in the collagen gel containing dispersed
CXCL12+HMFs show minimal invasion on day 5 of culture ...... 86
3.15: Co-culture spheroid model. (a) Normalized invasion of CXCR4+TNBC cells
from intermixed fibroblast-TNBC co-culture spheroids into collagen matrix.
(b) Representative fluorescent images of cancer cells spreading from
intermixed co-culture spheroids into the collagen matrix ...... 87
4.1: HGF-MET signaling pathway. The binding of hepatocyte growth factor
(HGF), the ligand of MET tyrosine kinase receptor (RTK), induces receptor
dimerization and phosphorylation of multifunctional docking site in the
kinase domain of the receptors. This activates many downstream signaling
pathways including MAPK, PI3K/Akt, and STAT3 ...... 121
4.2: CAFs-TNBC cells signal through HGF-MET pathway. (a) Phospho-RTK
arrays of MDA-MB-231 and SUM159 cells stimulated with conditioned xviii media of CAF-1 and CAF-2 monocultures. (b) Pixel densities representing
activities of EGFR and MET from treatments normalized with the respective
vehicle control and represented as activity fold change. Error bars represent
standard errors from a mean value. Data represent three separate
experiments. (c) Western blot analysis of effect of secretome of CAF-1 and
CAF-2 on MET activation in six TNBC cell lines. (d) Quantified levels of p-
MET normalizing with t-MET in TNBC cells stimulated with conditioned
media of CAF-1 and CAF-2 cultures and statistically compared with a
control group for each cell line. Data represent two separate experiments.
*p < 0.05 was calculated using two-tailed, unpaired t-tests. (e) The levels of
HGF, EGF, FGF-1, FGF-2, and PDGF AA/AB in the conditioned medium
derived from fibroblasts cultures were determined using bead based
multiplex immunoassay. CAF-1 and CAF-2 secreted HGF at significantly
higher levels than normal human mammary fibroblast (HMF) did. Other
soluble factors were not detectable. Data represent three separate
experiments. *p < 0.001 ...... 122
4.3: CAFs activate multiple oncogenic pathways in TNBC cells. (a) Western blot
analysis shows CAF-1 and CAF-2 activate pathways beyond those already
active, i.e., Akt and STAT3 in MDA-MB-231 cells and ERK1/2 and STAT3
in SUM159. (b) A MET inhibitor (crizotinib) downregulated oncogenic
signaling in MDA-MB-231 and SUM159 cell lines. The graph shows
quantified phospho-protein levels after normalizing with total protein levels
in TNBC cells when stimulated with conditioned medium from CAF-1 with xix or without MET inhibition (0.5 µM and 1 µM crizotinib). Statistical
comparison was done between TNBC cells stimulated with CAF-1
conditioned medium and the stimulated TNBC cells under crizotinib
treatment. Data represent two separate experiments. *p < 0.05 ...... 123
4.4: Met activity in EGFR+ TNBC. (a) Oncoprint of breast cancers (TCGA) shows
higher EGFR and MET gene alterations among basal and claudin-low
subtypes. (b) Analysis of METABRIC breast cancer database (TCGA)
shows that MET and EGFR expression is significantly higher in subtypes
representing TNBC disease. (c) MET expression strongly correlates with
EGFR expression both at a gene level (METABRIC breast cancer database
TCGA) and at a protein level (Firehose Legacy breast cancer database,
TCGA). (d) Analysis of functional proteomics data of tumors from patients
with invasive breast cancer derived from TCPA database shows a strong
correlation between active p-MET and p-EGFR. *p < 0.05 ...... 124
4.5: CAFs promote ECM invasion and proliferation of TNBC cells in a 3D
organotypic tumor model. (a) A two-step micropatterning approach to create
a tumor model that consists of a TNBC cell mass within a stroma composed
of ECM and dispersed CAFs. Confocal reconstruction of tumor model (blue:
TNBC mass; green: CAFs; collagen is not shown). (b) Confocal images of
TNBC cells on day 5 of culture show matrix invasion of MDA-MB-231 and
SUM159 cells promoted by CAFs, but not by HMFs. (c) Matrix invasion of
MDA-MB-231 and SUM159 cells when cultured with CAF-1 and CAF-2 and
xx normalized to the respective TNBC cells when cultured with HMFs. (d) A
typical flow cytometry result from organotypic tumor models with HMF, CAF-
1, or CAF-2, and TNBC cells. (e) Quantified flow cytometry results with
absolute counts of MDA-MB-231 and SUM159 cells. * p<0.05 ...... 125
4.6: CAFs promote EMT of TNBC cells. (a) CAFs promote a mesenchymal
morphology in TNBC cells in organotypic models. (b,c) CAFs increase
expression of EMT markers in TNBC cells both at gene and protein levels
(*p<0.05 compared to vehicle control (+HMF)) ...... 126
4.7: Screening of kinase inhibitors against TNBC spheroids. Dose−responses of
(a) MDA-MB-231 and (b) SUM159 spheroids to inhibitors of MAPK pathway
and PI3K pathway along with the list of molecular inhibitors used, their
targets, and LD50 values against MDA-MB-231 and SUM159 spheroids. The
‘–‘ symbol indicates an LD50 value could not be obtained ...... 127
4.8: Cyclic drug treatment and recovery of MDA-B-231:CAF-1 co-culture
spheroid. (a) The co-culture spheroids were cyclically treated with an
inhibitor of MEK (5 nM trametinib). (b) Kinetics of spheroids growth during
the cyclic treatment and recovery regimen. Each data point in the line graph
is an average of eight replicates. (c) Percentage viability of TNBC cells
measured from fluorescence intensity of endogenous GFP signal of TNBC
cells (n=8). (d) Growth rate (kc) of co-culture spheroids during four treatment
rounds with trametinib (n=8); *p<0.01. Error bars in panels (b-d) represent
the standard errors from a mean value ...... 128 xxi 4.9: Targeting CAFs-mediated MDA-MB-231 pro-metastatic functions. (a) A
typical matrix-format dose-dependent combination treatment with crizotinib
(MET inhibitor) and trametinib (MEK1/2 inhibitor) to block matrix invasion of
MDA-MB-231 cells. A combination concentration (red box) that significantly
blocks invasiveness with a high synergy (CI<0.5) is shown. (b) Activity
levels of ERK, Akt, and STAT3 in MDA-MB-231 organotypic cultures
following combination treatments with 1 µM crizotinib and 5 nM trametinib.
(c) Quantified levels of p-ERK/t-ERK, p-Akt/t-Akt, and p-STAT3/t-STAT3
followed by single-agent or combination treatments; *p<0.05. Data
represent two separate experiments. (d) Quantified flow cytometry results
with absolute counts of MDA-MB-231 from free-floating spheroids (MDA-
MB-231 only) and organotypic cultures containing CAF-1 (control), single-
agent treatments (+1 µM crizotinib or +5 nM trametinib) and combination
treatment (+crizotinib+trametinib); * p<0.05 and n=5. (e) mRNA fold change
values of seven EMT transcription factors before and after single-agent and
combination treatments. Two-tailed, unpaired t-test was used to calculate
the p-values; * p<0.05 ...... 129
4.10: Targeting CAFs-mediated SUM159 pro-metastatic functions. (a) A typical
matrix-format dose-dependent combination treatment with crizotinib (MET
inhibitor) and dactolisib (PI3K inhibitor) to block matrix invasion of SUM159
cells. A combination concentration (red box) that significantly blocks
invasiveness with a high synergy (CI<0.5) is shown. (b) Activity levels of
ERK, Akt, and STAT3 in SUM159 organotypic cultures following xxii combination treatments with 1 µM crizotinib and 1 µM dactolisib. (c)
Quantified levels of p-ERK/t-ERK, p-Akt/t-Akt, and p-STAT3/t-STAT3
followed by single-agent or combination treatments; *p<0.05. Data
represent two separate experiments. (d) Quantified flow cytometry results
with absolute counts of SUM159 from free-floating spheroids (SUM159
only) and organotypic cultures containing CAF-1 (control), single-agent
treatments (+1 µM crizotinib or +1 µM dactolisib) and combination treatment
(+crizotinib+dactolisib); * p<0.05 and n=5. (e) mRNA fold change values of
seven EMT transcription factors before and after single-agent and
combination treatments. Two-tailed, unpaired t-test was used to calculate
the p-values; * p<0.05 ...... 130
4.11: HGF-MET signaling effect in a lung stromal environment. (a) MET signaling
is enriched in murine breast cancer with lung metastases (GSE138139).
The GSEA enrichment chart of MET signature genes in a murine model with
lung metastases compared to parental TNBC cells. (b) Normal lung
fibroblasts (WI-38) activated MET in TNBC cells. (c) WI-38 cells produced
approximately 7.2 ng/ml of HGF. (d) WI-38 cells promoted formation of
TNBC colonies in the lung stromal environment, and crizotinib significantly
suppressed clone formation of SUM159, Hs578T, and MDA-MB-157 TNBC
cells. (e) Immunofluorescence staining of Ki-67 (red) in SUM159, Hs578T,
and MDA-MB-157 colonies with and without WI-38 cells or crizotinib. Green
cells represent TNBC cells. Blot plots of normalized Ki-67+ cell count per
spheroid area for three different conditions. The boxes represent the 25th xxiii and 75th percentiles with the median shown with a horizontal line inside
each box. The mean is shown with a cross symbol. The whiskers represent
the 10th and 90th percentile of the data. Two-tailed Mann-Whitney test was
used to calculate p-value; *p<0.01 and n=7 ...... 131
4.12: Confocal images of organotypic cultures of MDA-MB-231 and SUM159
spheroids on day 5 of culture with and without inhibitors (1 µM crizotinib,
neratinib, or lapatinib). The graphs show normalized invasion area of MDA-
MB-231 and SUM159 cells with that of vehicle control; *p<0.01 and n=6
...... 132
A-1: MET RTK activated by TNBC-CAFs interaction. Phospho-RTK arrays of (a)
MDA-MB-231 and (b) SUM159 cells stimulated with conditioned media of
CAF-1 and CAF-2 monocultures showing pixel integrated densities
representing activities of several RTKs from treatments normalized with the
respective vehicle control and represented as activity fold change. Error
bars represent standard errors from a mean value. Data represent two
separate experiments...... 174
xxiv LIST OF TABLES
Table Page
1.1: Advantages and disadvantages of different preclinical tumor models. .... 21
4.1: List and sequences of primers of the genes analyzed for EMT of TNBC
cells...... 120
xxv CHAPTER I
INTRODUCTION
Portions of this chapter are reused from:
S. Singh, S. Tran, J. Putman, H. Tavana, Three-dimensional models of breast cancer–fibroblasts interactions, Exp. Biol. Med. 245 (2020) 879–888. © 2020, SAGE publications. https://journals.sagepub.com/doi/abs/10.1177/1535370220917366
1.1 Significance
Cancer is a major public health problem and the second leading cause of death in the United States [1]. The National Cancer Institute estimated 1,762,450 new cases of cancer and 606,880 cancer deaths in the United States in 2019.
Cancer survival has significantly improved for many cancers since the mid-1970s, partly due to earlier diagnosis, improvements in identifying molecular drivers of the disease to guide therapies, and discovery of less toxic and more effective targeted therapies than conventional cytotoxic chemotherapies. Breast cancer is the most diagnosed cancer and the second leading cause of death among women in the
United States after lung cancer [2,3]. While the breast cancer death rate continued to decline by 40% from 1989 to 2017, the death rate is 40% higher in African- 1
Americans despite a lower incidence rate [3]. This underlines an important problem of cancer health disparity among patients of different ethnicities. The death rate from breast cancer has been declining since the early 1900s mainly due to earlier detection of palpable tumors, mammography screening, and adjuvant systemic therapy [4].
Cancer was considered as a single disease where mutations of normal cells led to generation of malignant cells. It is now widely recognized that cancer is a disease that involves not only cancer cells, but also different cells and proteins in the tumor stroma including fibroblasts, infiltrating immune cells, blood vessel cells, the extracellular matrix (ECM), growth factors, chemokines, and cytokines [5,6].
As cancer progresses, the stromal components co-evolve with cancer cells to form a reactive stroma that promotes cancer growth, invasiveness, and metastasis of tumor cells [7]. Various components of tumors and their dynamic interactions make up the tumor microenvironment (TME). Due to the direct role of the TME in cancer progression, therapy approaches to target both cancer cells and the tumor stroma may offer better outcomes for patients [8]. Traditional chemotherapies for cancer are largely ineffective due to drug resistance [9]. But unlike cancer cells, the stromal cells are comparatively genetically stable and less likely to be resistant to drugs, indicating a benefit of targeting the tumor stroma [10,11]. This is especially more important for cancers that have very limited targeted therapy options, such as triple negative breast cancer and pancreatic cancer, underscoring the need for discovery of novel therapeutics against the tumor-stromal interactions.
2 1.2 Triple Negative Breast Cancer (TNBC)
TNBC refers to a subtype of breast cancer that lacks expression of estrogen receptor (ER) and progesterone receptor (PR), and amplification of HER2 receptor
(ERBB2). Other subtypes of breast cancer include luminal A, luminal B, and HER2
[12]. Lumina A tumors are ER/PR positive and often low-grade and effectively treated with hormonal therapies [13]. Luminal B tumors are ER positive, contain higher proliferative Ki67+ cells, are more aggressive than the luminal A subtype, and often relapse. HER2 tumors are most common but are more successfully treatable [14]. TNBC accounts for ~15% of all invasive breast cancers but is the most aggressive breast cancer and claims disproportionately more lives. TNBC is a heterogeneous disease composed of at least 4 different molecular subtypes: basal like, mesenchymal, immunomodulatory, and androgen receptor-positive, each with its own unique molecular characteristics [12,15]. The median survival of patients with metastatic TNBC is only 13.3 months and most of the patients with metastatic TNBC will eventually die despite systemic therapy [16]. Current therapies for breast cancer include surgery, chemotherapy, radiation, endocrine therapy, and targeted therapy. However, there are very limited targeted therapy options for TNBC. Unfortunately, endocrine and hormonal therapies are not feasible with TNBC. Additionally, TNBC is less responsive to conventional chemotherapy that other breast cancers [17]. Therefore, there is an unmet need for new treatment strategies for TNBC. Recent evidence suggests that the TME and its interactions with cancer cells (i.e., tumor-stromal interactions) confer cancer cells with drug resistance and enable metastatic tumor progression [11,18]. 3 Advanced TNBC tumors are rich in stroma [19], indicating that targeting the tumor stroma is rational therapeutic approach.
1.3 Tumor Microenvironment (TME)
The TME is a complex tissue containing cancer cells and various stromal cells such as cancer-associated fibroblasts (CAFs), immune cells, and endothelial cells embedded in an extracellular matrix (ECM) network [20]. Interactions among the components of the tumor milieu create the TME. The stromal cells create a niche that dynamically supports the tumor cells. The tumor-stromal interactions are mainly driven by dynamic networks of growth factors, chemokines, cytokines, and enzymes produced by both cancer and stromal cells. Epigenetic alterations of cancer cells, driven by the tumor stroma, lead to aberrant gene expression in cancer cells, and stromal gene signatures are a predictor of cancer prognosis [21].
The breast TME can be discussed based on the location of the cancer cells: local (primary) or distant (metastatic) TME [22]. In normal mammary tissue, stromal cells such as fibroblasts have essential roles in tissue homeostasis. However, as cancer progresses, the stromal cells become activated through signaling cues of cancer cells. The activated stromal cells secrete soluble signaling molecules that promote proliferation of cancer cells and tumor growth. As the tumor grows and its metabolic needs increase, cancer cells recruit the nearby vessels to sprout capillaries into the tumor and provide nutrients for cancer cells. These capillaries also provide a route for the escape of cancer cells from the primary tumor. A small number of cancer cells separate from the tumor mass, migrate and invade through 4 the ECM, reach the capillaries, intravasate to access the circulation, and eventually metastasize to distant organs [23].
The circulating tumor cells that reach a distant organ often enter a dormant phase that can last months to years before forming metastases. The interactions between the disseminated cancer cells and the host microenvironment is an important determinant of successful formation of metastatic tumors. TNBC cells often metastasize to lungs, brain, and the skeletal tissue [24,25]. The cancer cells have plasticity in that they can adapt to the specific metastatic microenvironment to survive and proliferate. For example, cancer cells interact with osteoblasts, osteoclasts, and hematopoietic stem cells in bones but with lung fibroblasts in lungs [23]. Altogether, the stroma is critical to progression of cancers in both primary and metastatic sites.
1.3.1 Components of TME
The major components of the TME that have pro-tumorigenic functions are discussed below.
- Extracellular matrix (ECM)
The ECM is a complex network of proteins and provides structural support for cells, enables interactions among the components of the TME, and directly interacts with cancer cells by providing cell-binding sites. The ECM consists of proteins such as collagen, elastin, fibronectin, hyaluronic acid, laminin, and proteoglycans [26]. Large amounts of ECM proteins in tumors result in the stiffening of the tumor stroma. In the past, ECM was considered to be a passive 5 component of the TME but research in the past two decades has shown that ECM provides biochemical and biophysical signaling cues to cancer cells to facilitate tumor progression [27]. Various studies have shown that a stiffer ECM correlates with invasiveness and malignant behavior of cancer cells [28–30]. During cancer progression, ECM undergoes continuous remodeling that results in structural and mechanical changes. This reorganization of the ECM promotes tumor progression by disturbing cell polarity, altering integrin adhesions, and inducing focal adhesions
[28,31,32]. Tumor stiffness is considered a prognostic factor and stiffer tumors often have a poorer prognosis [33,34].
-Fibroblasts
Among different stromal cells, fibroblasts play pivotal roles in initiation and progression of epithelial tumors. In normal tissues, fibroblasts are present in the stroma, deposit ECM proteins such as collagen and fibronectin [35], and degrade the ECM by producing matrix metalloproteinases (MMPs) [36]. Balanced remodeling of the ECM is necessary for normal functioning of tissues, whereas disruption in the remodeling process leads to pathologies such as fibrosis and cancer [37]. Fibroblasts are also important in wound healing. Fibroblasts migrate into the lesions and transform into myofibroblasts to accelerate ECM deposition.
Myofibroblasts are activated fibroblasts with a unique spindle shape and express
α-smooth muscle actin (α-SMA) and fibroblast activation protein (FAP) [38,39].
After the wound healing is complete, the myofibroblasts may revert back to a normal phenotype or undergo apoptosis [40]. In tumor tissues, normal fibroblasts are recruited in large numbers by cancer cells to convert the TME into a reactive
6 stroma [41]. Fibroblasts are transformed into an activated phenotype by various stimuli such as transforming growth factor-β (TGF-β) and epidermal growth factor
(EGF). Activated fibroblasts in the TME are called cancer-associated fibroblasts
(CAFs). CAFs are activated permanently and do not revert to a normal phenotype nor do they undergo apoptosis [42].
CAFs are the most abundant stromal cells in solid tumors and associate with cancer cells during tumor initiation, progression, invasion, and metastasis [43].
Immunohistochemical analysis of human tumors showed that abundance of CAFs correlates with a poor prognosis [44]. CAFs are derived from various cell types such as resident fibroblasts, vascular smooth muscle cells, endothelial cells, and pericytes, and hence are heterogeneous in origin [45,46]. CAFs support tumor growth through soluble signaling of various CAFs-derived growth factors such as
TGF-β [47], hepatocyte growth factor (HGF) [48,49], fibroblast growth factor (FGF)
[50], interleukin (IL-6) [51,52], stromal derived factor-1α (SDF-1α or CXCL12)
[53,54], and EGF [55] (Figure 1.1). This recognition has prompted a therapeutic strategy using inhibitors of the corresponding receptors, i.e., EGFR [56,57], MET
[58], FGFR [59], VEGFR [60], and CXCR4 [61,62], that are expressed on cancer cells to inhibit the oncogenic signaling pathways downstream of these receptors.
-Immune cells
When cancer progresses, stromal cells including immune cells are recruited in and around the tumor tissue in an attempt to eradicate the neoplastic cells [63].
Infiltration of innate immune systems like nature killer (NK) cells has been associated with a favorable prognosis. However, infiltration of other immune cell
7 types such as macrophages and mast cells has been associated with unfavorable prognosis [64]. Although the function of immune cells is to eliminate cancer cells, the recruited immune cells can change their phenotypes and functions to provide a tumorigenic niche for cancer progression. For example, macrophages convert into tumor-associated macrophages (TAMs). Animal models studies show that
TAMs promote tumor cell dissemination through angiogenesis and invasion
[65,66]. Cancer cells evade the immune system mainly by secretion of immunosuppressive factors such as IL-10, TGF-β, and VEGF and through expression of immune checkpoint ligands such as program death-ligand-1 (PD-
L1). CAFs have also been reported to suppress cancer killing capability of T cells by secreting fibroblast activation protein (FAP) [67]. There are also reports that support a tumor promoting role of B lymphocytes [68]. Novel cancer immunotherapies seek to leverage the cancer-eradicating intrinsic capability of immune cells by inhibiting an immunosuppressive TME.
-Other components
Formation of new vasculatures in solid tumors to promote tumor growth and metastasis depends on endothelial cells. These cells in the TME also transform into fibroblast-like cells with increased expression of α-SMA [69]. Transformed endothelial cells may express a normal endothelial cell marker, CD31, but contain genetic abnormalities with aneuploid cells and abnormal centrosomes [70]. These cells have been found to separate from primary tumors and co-migrate with tumor cells in the blood stream to distal sites and protect the circulating tumor cells from anoikis and apoptosis [71,72].
8 Pericytes are perivascular cells involved in vasculature development along with endothelial cells. These cells secrete growth factors that stimulate endothelial cells and help with the tumor angiogenesis process (formation of capillaries).
Pericytes are a negative regulator of metastasis and a low number of pericytes in the tumor vasculature correlates with a poor prognosis and increased metastasis
[73]. Pericytes also have been shown to have immunosuppressive properties within the TME by reducing the proliferation of cytotoxic T lymphocytes [74].
The TME contains several other cellular and non-cellular components that may contribute to the tumor initiation and progression, as discussed in several comprehensive reviews [75–77].
1.3.2 Tumor Stroma as Therapeutic Target
Compelling evidence suggests that targeting cancer cells alone is not sufficient for treating cancer patients because cancer cells often develop resistant to conventional therapies both intrinsically and through interactions with the tumor stroma [11], i.e., fibroblast-mediated [18,78], ECM-mediated [79,80], vascular- mediated [81], and immune-mediated resistance [82]. Therefore, targeting the
TME and tumor-stromal interactions are critical to improve outcomes for patients.
Strategies to target tumor stroma can be summarized into two categories: target stromal components (such as ECM, CAFs, and immune cells), and interfere with growth factor/chemokine/cytokine-mediated tumor-stromal interactions [83].
Several strategies to reduce lysyl oxidase (LOX)-dependent crosslinking of collagen fibers have been shown to reduce tumor fibrosis and metastases [84]. 9 Inhibition of MMPs is a promising therapeutic approach since they are important in
ECM remodeling and providing a pro-invasive niche for cancer cells [85]. Because of the importance of CAFs in tumor progression, CAFs provide an ideal pharmacological target. FAP is expressed on the surface of CAFs and can be used as a novel target. Signaling between CAFs and cancer cells occurs through various receptors on cancer cells. Receptor tyrosine kinases (RTKs) are expressed on
TNBC cells and are activated by ligands often produced by stromal cells. Several drugs against these interactions are under various stages of clinical trials, including against VEGFR, PDGFR, MET, FGFR, and EGFR [86]. Other therapeutic strategies include targeting specific signaling pathways induced by CAFs-tumor cells interactions such as SDF1-CXCR4, notch-hedgehog, and wnt/β-catenin pathways [87,88], and inhibition of downstream pathways such as MAPK, PI3K,
JAK/STAT pathways [12]. Anti-VEGF treatments have also significantly improved overall survival in colorectal, lung, and breast cancer patients [89]. Additionally immunotherapy approach such as anti-PD-1 has shown promising clinical results
[90,91]. Anti-PD-1, atezolizumab, has recently been approved by FDA in combination with chemotherapy for advanced TNBC [92].
While much progress has been made in understanding the role of TME in carcinogenesis and development of compounds that target tumor stroma, the success rate of anti-cancer compounds in clinical trials is low [93]. It is widely accepted that the lack of predictive and biologically relevant preclinical models to sort out effective and ineffective compounds in preclinical studies is a major impediment to improve the success rate of cancer treatments [94]. As such,
10 developing physiologically relevant preclinical tumor models is currently pursued to facilitate drug discovery against tumor-stromal interactions.
1.4 Models of Tumor-Stromal Interactions
Considering the importance of stromal cells in tumor progression, a mechanistic understanding of interactions among components of the TME is critical to develop novel therapeutic strategies. Typically, animal models and in vitro 2D cell-based assays are extensively used in preclinical studies to study tumor-stromal interactions and assess drug efficacy and safety. In recent years,
3D cultures have gained popularity because they enable adjusting matrix composition and cellular components to develop a more physiologic model.
Several advantages and disadvantages of preclinical tumor models are listed in
Table 1.1.
1.4.1 2D Cell Culture Models
A vast majority of ex vivo studies are performed in 2D cultures in plastic
Petri dishes and plates. Monolayer cultures are convenient to use, adaptable with robotic instruments used in the pharmaceutical industry, allow high throughput screening of chemical compounds, and enable straightforward analysis of responses of cells to drug compounds. In vitro co-cultures of cancer cells and stromal cells such as fibroblasts in a monolayer have also been widely used to study stroma effect on phenotypes and functions of breast cancer cells such as proliferation, invasion, and apoptosis [95–97]. However, there are several 11 limitations associated with 2D cell cultures. These include lacking cell-cell interactions, cell-stroma interactions, and the 3D geometry and architecture of human tumors. Drugs which are toxic in 2D cultures often display reduced toxicity in 3D cultures of cancer cells. Importantly, a study showed no correlation between multidrug resistance (MDR) transcriptome of cancer cell lines grown in 2D culture with clinical samples of their corresponding cancer types [98]. Cells grown in 2D cultures also had different metabolic and gene expression profiles than in 3D cultures [99]. The literature contains a large amount of information on the disparity between 2D cultures and native tumors in terms of cell functions and therapy responses, indicating a need for more physiologic models.
1.4.2 Animal Models
Animal models, in particular mouse models, of human cancer have proved as a useful research tool due to their anatomical and physiological similarities with humans. Mouse models have played a significant role in understanding mechanisms of tumor development, angiogenesis, invasion, and metastasis, and identifying novel biomarkers for testing the efficacy of cancer drugs [100–103].
Many mouse models have been widely used to study effects of stromal CAFs on breast cancer progression [104–106]. Currently there are improved mouse models such as patient-derived xenografts (PDX) and genetically engineered mouse models (GEMM) for preclinical therapy testing and to model disease subtypes and metastasis [107]. Nevertheless, limitations of mouse models including significant failure rates in developing tumors in mice, differences in the tumor stroma and 12 immune system between mouse and human, difficulty of handling a large number of animals and analysis of responses to treatments, low throughput for drug screening applications, expense, and ethical issues remain major obstacles to accelerate new discoveries [108,109]. Additionally, it is difficult to decipher role of a specific stromal cell or a specific biological cue that results in changes in cancer cell behavior in a mouse model due to presence of different stromal cells and other confounding factors [110].
1.4.3 3D Cell Culture Models
The shortcomings of 2D and animal models emphasizes an immediate need for novel preclinical models to understand the biology and mechanisms of tumor- stromal interactions [111,112]. 3D culture technologies offer significant advantages over 2D models and provide a physiologic tool to recreate the complexity of the TME. 3D tumor models can be broadly classified into three categories based on the technologies used to develop them: scaffold-free systems, scaffold-based systems, and tumor-on-chip systems [113,114].
- Scaffold-free systems
Culture of cells as spheroids represents the scaffold-free systems.
Spheroids are 3D compact aggregates of cancer cells, or cancer and stromal cells, that freely float in culture media. Studies show that spheroids reproduce certain aspects of solid tumors including close cell-cell contacts, spatial gradients of nutrients and oxygen, hypoxia and necrosis, expression of pro-angiogenic proteins, and multidrug resistance of cancers [115–117]. Although spheroids can 13 be made exclusively of cancer cells [118], including multiple cells such as cancer and stromal cells allows studies of tumor-stromal interactions and effect of stromal cells in cancer proliferation, progression, and therapeutic responses [119–123].
Several methods have been developed over the years to generate spheroids, such as spinner flasks [124], hanging drop [125], magnetic levitation
[126], ultra-low attachment plates [127], and aqueous two-phase system (ATPS)
[128,129]. Spinner flasks or bioreactors are a common method for large-scale production of tumor spheroids where cell aggregates are stirred in a suspension culture. Drawbacks of this technology is that the cells experience sheer force generated by continuous motion of the stirring bar, which may affect the cellular physiology, and size heterogeneity of spheroids [125]. The traditional hanging drop method uses cancer cells suspended in a pendant drop to produce spheroids. This technique produces relatively uniformly-sized spheroids but is associated with difficulty of handling plates during culture and need to transfer spheroids to a standard well plate for subsequent cell-based assays, making it less appealing
[113]. Another technique called magnetic levitation utilizes paramagnetic iron oxide nanoparticles to guide self-assembly of cells into spheroids under magnetic forces. Ultra-low attachment (ULA) plates use hydrogel formulations to prevent cell attachment to the well surface and promote formation of reproducibly-sized spheroids in high throughput [130]. In addition to difficulty of making spheroids with specific cell types, spheroids made with ULA plates tend to be disk-shaped than spherical [131,132]. A technique developed in our laboratory used an aqueous two-phase system (ATPS) for spheroid formation and overcomes the difficulties of
14 other methods. This technology conveniently enables forming consistently-sized spheroids and have been used extensively to study tumor-stromal interactions and to model cancer drug resistance [115,133].
- ATPS for spheroid formation
ATPS consists of two immiscible aqueous solutions of chemically incompatible polymers such as polyethylene glycol (PEG) and dextran (DEX).
Each ATPS has a characteristic phase diagram with a binodal curve that prescribes pairs of concentrations of its pair of polymers to result in two immiscible phases (Figure 1.2). Only polymer concentrations above the binodal curve give a two-phase system. ATPS has been used for separation of biomolecules such as cells, proteins, and nucleic acids [134–136]. Partitioning of proteins to either phase is a complex process and depends on factors such as pH, ionic strength of aqueous solution, hydrophobicity of the polymers and molecular weight of the polymers [137,138]. ATPS provides a favorable environment for various cells like cancer cells to grow as a spheroid. Utilizing partitioning of cells in ATPS, cancer cells mixed with aqueous DEX phase solution is dispensed as submicroliter droplet into a non-adherent microwell containing aqueous PEG phase solution. The cells are entrapped in the denser aqueous DEX drop and spontaneously aggregate to form a spheroid. ATPS have ultralow interfacial tensions, on the scale of 0.001 –
0.01 mJm-2 between the two aqueous phases, which results in effective partitioning of cancer cells to the DEX phase drop and formation of a spheroid [139]. This technique has been used to form co-culture spheroid models of TNBC and CAFs
15 to study tumor-stromal signaling, cancer proliferation, and matrix invasion of cancer cells [133].
- Scaffold-based systems
Scaffold-based systems provide an ideal environment to mimic the native
3D architecture. A scaffold resembles the ECM in tumors, provides structural support for the cellular components, and enables cell-matrix interactions. Several studies have used 3D scaffolds to study effects of stromal cells on proliferation, invasion, metastasis, and immunosuppression of breast tumor cells [140–143].
These scaffolds are either derived from natural materials (e.g., collagen I, collagen
IV, and Matrigel) or synthetic materials (such as PEG, poly(d,l-lactic-co-glycolic)
(PLGA), and polylactide (PLA)) [114,144].
Stromal fibroblasts induced invasive growth of squamous cell carcinoma by upregulation of MMP-2 and MMP-9 in a collagen I-based hydrogel tumor model
[145]. A collagen IV hydrogel was used to study the role of fibroblasts on chemosensitivity of breast cancer cells [146]. Matrigel is an animal-derived protein cocktail containing different ECM components such as laminin, collagen type IV, proteoglycans, as well as growth factors such as TGF-β and FGF [147]. CAFs co- cultured with murine liver tumor organoids in Matrigel promoted organoid growth and conferred resistance to anticancer drugs [148]. Co-culture of murine organoids with mesenchymal cell-derived CAFs in a Matrigel facilitates tumor-stromal interactions mediated by CAFs-secreted IL-6 [149]. However, the pitfall of using
Matrigel is that it is derived from animal tumor, contains many growth factors not necessarily present in human tumors, does not allow adjusting the ECM
16 composition and its mechanical properties for mechanistic studies of cell-ECM signaling, and may interfere with the effect of stromal cells due to presence of basement membrane and growth factors [150,151]. A more general drawback of natural scaffolds is the batch-to-batch variability of the materials used.
Synthetic materials are also used to form scaffolds. These materials can be biodegradable, and their properties can be tailored according to the application.
Although synthetic scaffolds are mechanically stronger than natural ECMs, they need to be chemically functionalized to promote cell adhesion and growth. PEG hydrogels were functionalized with peptides (RGD sequences) to promote cell-cell and cell-ECM interactions [152]. PEG-derived hydrogels were used to demonstrate a matrix stiffness dependent tumor-stromal interactions and stromal cell adipogenesis [153]. Although synthetic scaffolds are easily available and provide a wide range of tunable material properties, their major drawbacks are lack of native ECM proteins, difficulty in imaging due to thickness and non-transparency of the materials, potential cytotoxicity from chemicals used for their fabrication, and limited bioactivity [114,154,155].
- Tumor-on-chip systems
Microfluidic systems enable addition of various components of the TME to study cell-cell and cell-matrix interactions in a 3D environment. These models provide control over matrix composition, its mechanical properties, and spatiotemporal composition of cellular components under physiological flow conditions. A microfluidic device was developed to study dispersion of breast cancer cells due to chemokines secreted by fibroblasts [156]. Recently a 3D co-
17 culture invasion assay was used to study cross-talk between breast cancer cells and fibroblasts by mimicking the spatial organization of the tumor microenvironment on a chip [157]. Another study used a compartmentalized microfluidic system and showed the role of fibroblasts on the transition of ductal carcinoma in situ (DCIS) to invasive ductal carcinoma (IDC) mediated by paracrine signaling and cell-cell contact signaling [158]. ECM remodeling during progression of cancer was studied in vitro using a tumor-on-chip model that recapitulates stromal activation during tumor cell invasion [159]. Transient interactions between endothelial cells and cancer cells mediated by CXCL12-CXCR4 chemokine- receptor signaling during metastasis was also investigated to demonstrate the feasibility of mechanistic studies [160]. Although microfluidic systems provide great flexibility of device design and control over biophysical and cellular components, they have several drawbacks such as complex chip design, need for highly experienced users to form and maintain microfluidic cell cultures, incompatibility with industrial-scale high throughput drug screening applications, and adsorption of drugs and proteins on the surface of polydimethylsiloxane (PDMS) devices
[161]. Recently there have been efforts to develop microfluidic devices that are compatible with high throughput drug screening [162,163]. However, there are several issues such as leakage across compartments and difficulty to perform multi-step liquid handling [163].
As briefly discussed above, current models of tumor-stromal interactions inadequately represent the 3D native tumor environments and its constituents for mechanistic studies, are not compatible with automated high-throughput drug
18 screening and may need expertise of highly trained users. Therefore, there is currently an unmet need for physiologic tumor models that overcomes these issues to expedite development of effective treatments.
1.5 Aims and Scope
Through this doctoral dissertation, we establish a broad utility of our ATPS technology to develop physiological relevant 3D organotypic cultures by leveraging cell and protein partitioning in ATPS. The 3D organotypic cultures enable mechanistic studies of tumor-stromal interactions to facilitate design and testing of new treatments. The main objective of this work is to (i) develop and characterize an organotypic breast tumor model, (ii) biologically validate the organotypic breast tumor model, and (iii) establish the feasibility of targeting tumor-stromal interactions as a therapy approach. The principle of micropatterning organotypic cultures using cell and protein partitioning in ATPS will be presented in Chapter II.
This work will streamline formation of tumor models that recapitulate the spatial distribution of cells in their native environments and reproduce mechanical properties of breast tumor. The tumor model formation is compatible with high throughput setting to enable study of tumor-stromal interactions to design treatment modalities and test them. In Chapter III, we will biologically validate the organotypic breast tumor model using a well-known signaling axis that involves a fibroblast cell-secreted chemokine CXCL12 and its CXCR4 receptor on TNBC cells to recapitulate a major hallmark of cancer, i.e., cancer cell invasion. We will demonstrate the biological relevance of our tumor model by recapitulating key 19 tumor biological processes such as matrix remodeling and cancer cell invasion of the ECM. In Chapter IV, we will explore and identify prominent CAFs-derived signaling molecules to design treatment strategies that inhibit interactions of CAFs and TNBC cells. Importantly, we will demonstrate that TNBC cells develop resistance to single-agent therapies to justify a need for combination drug treatment strategies to suppress pro-metastatic events in TNBC driven by interactions with CAFs. These chapters will collectively demonstrate the utility of
ATPS technology to generate a physiologically relevant breast tumor model, elucidate specific mechanisms of tumor-stromal interactions, and develop stroma- targeted therapies. Major conclusions of this dissertation and future work are discussed in Chapters V and VI, respectively.
20 Table 1.1: Advantages and disadvantages of different preclinical tumor models.
Models Advantages Disadvantages
+ Convenient to use and amenable to -Lack cell-cell and cell-ECM imaging interactions 2D + Allow high throughput drug screening -Lack 3D architecture of native + Inexpensive culture method tumor -Lack diffusion gradients + In vivo like 3D architecture -Lack vasculatures + Enable adjusting matrix compositions -Short-term studies 3D and cellular components -Difficult to adapt to high + Diffusion gradients of oxygen, drugs, throughput screening and hypoxia + Complex heterogenic tumor -Expensive microenvironment -Time consuming Animal + Allow long-term studies - Lack human stroma + Whole living organism and provide systemic drug response
21
Figure 1.1: Cancer cells secrete soluble factors to activate fibroblasts into cancer- associated fibroblasts (CAFs), which in turn remodel the tumor microenvironment and promote tumor growth, invasion, and metastasis.
22
Figure 1.2: Phase diagram of ATPS with 35 kDa PEG and 500 kDa DEX. The solid line represents binodal curve obtained by fitting experimental data (open circles). The dashed line represents tie-line to show the locations of initial and final compositions of an ATPS. Combinations of concentrations of PEG and DEX above the binodal curve will result in two phases. The inset cartoon shows equilibrated ATPS separated into two distinct PEG-rich and DEX-rich phases (Adapted with permission from E. Atefi, J. A. Mann, and H. Tavana, “Ultralow interfacial tensions of aqueous two-phase systems measured using drop shape,” Langmuir, vol. 30, no. 32, pp. 9691–9699, 2014. Copyright © 2014 American Chemical Society) [164].
23
CHAPTER II
COLLAGEN PARTITION IN POLYMERIC AQUEOUS TWO-PHASE
SYSTEMS FOR TISSUE ENGINEERING
Portions of this chapter are reused from:
S. Singh, H. Tavana, Collagen Partition in Polymeric Aqueous Two-Phase Systems for Tissue Engineering, Front. Chem. 6 (2018). © 2018 Singh and Tavana. Reprinted through open-access terms of the Creative Commons Attribution License (CC BY). https://pubmed.ncbi.nlm.nih.gov/30234101/
Heterogeneous tumor microenvironment (TME) consists of several cell types, matrix proteins, and soluble signaling molecules, and interactions among these components play critical roles in cancer pathogenesis [111]. There has been increased interest in using in vitro cell culture platforms to identify molecular and cellular mechanisms by which the TME promotes cancer progression and regulates therapy responses [111]. Traditional two-dimensional (2D) cultures and animal models do not adequately represent human tumors [108,109]. Therefore, three-dimensional (3D) models that closely mimic biophysical and biological properties of solid tumors have gained immense popularity among the research
24 community [113]. 3D tumor models have primarily been based on spheroids [129], microfluidics [165], and cell encapsulation in natural or synthetic scaffolds [166].
Some of the drawbacks of existing 3D models are incompatibility with automated high throughput systems, labor-intensive formulation of scaffold-based models, lack of key components of the TME, and architectural differences with native tumors. Although development of organoids from patient tumor specimens and from stem cells has been a major improvement over existing 3D cultures, poorly- defined basement membrane matrix, formation of organoids takes a long time
(often up to 3 months) and is not always successful [167,168]. The objective of this chapter is to demonstrate a cell and protein micropatterning approach using aqueous two-phase systems (ATPS) toward developing a 3D tumor model of human breast tumor that overcomes the shortcomings of currently-used tumor models.
Aqueous two-phase systems (ATPS) may be formed by mixing aqueous solutions of two chemically incompatible polymers [169]. Polyethylene glycol
(PEG) and dextran (DEX) are the most common polymers to use for ATPS formation. Each ATPS has a characteristic phase diagram with a binodal curve that prescribes pairs of concentrations of its pair of polymers to result in two immiscible phases. Only polymer concentrations above the binodal curve give a two-phase system. Separation of two distinct aqueous layers by an interface is visible and becomes more distinct by increase in the interfacial tension between these two phases [139,164]. ATPS are used for separation of biomolecules such
25
as cells [135] , proteins [134], nucleic acids [136], and organelles [170]. Unlike two- phase systems made with water and oil or organic compounds, high water content
(usually > 90 %w/w) and ultralow interfacial tensions of ATPS are key properties to provide a mild environment for sensitive biomolecules. Partitioning of proteins to either phase of an ATPS is a complex process [137]. Proteins may favor one of the phases of an ATPS or partition toward the interface. Distribution of protein molecules may be manipulated by altering the molecular weight of the polymers, concentration of polymers, ionic strength of the aqueous solution, pH, and hydrophobicity of the polymers [138]. For example, amylase partitioning to the top phase improved by adding salt to the aqueous solutions [171]. Using charged PEG increased partitioning of penicillin acylase from E. coli to the top phase in an ATPS
[172]. It was recently shown that collagen partitions to the interface of aqueous
PEG and DEX phases [173]. This was used to print cell-containing collagen microdrops to resemble matrix contraction.
Our goal was to localize collagen to one phase, rather than to the interface, to conveniently form collagen microgels in ATPS. To demonstrate the feasibility of partitioning of collagen to the bottom DEX phase, we conducted experiments with two-phase systems of different molecular weights of PEG and DEX. Then, we selected a system that favors partition of collagen to the DEX phase and used this property to develop microtissues that resemble solid breast tumors. This novel approach may enable future studies in tumor biology and antitumor drug discovery.
26
2.1 Materials and Methods
The methods for preparation of aqueous two-phase, construction of binodal curves, partition of collagen in ATPS, and embedding cancer cell spheroids in partitioned collagen are described below.
2.1.1 Preparation of Aqueous Two-Phase Systems (ATPS)
ATPS were prepared with polyethylene glycol (PEG) (Sigma) and dextran
(DEX) (Pharmacosmos). Two different molecular weights of PEG (8 kDa & 35 kDa) and DEX (40 kDa & 500 kDa) were used to prepare four ATPS: PEG 8 kDa-DEX
40 kDa (system A), PEG 8 kDa-DEX 500 kDa (system B), PEG 35 kDa-DEX 40 kDa (system C), and PEG 35 kDa-DEX 500 kDa (system D). From each system, different stock concentrations of PEG and DEX solutions were used to form three sets of two-phase systems: 15% PEG-21% DEX, 18% PEG-24% DEX, and 24%
PEG-32% DEX. All concentrations of aqueous PEG and DEX solutions were calculated in %(w/v). Polymers were dissolved in a complete growth medium. To facilitate complete dissolution of polymers, the solutions were kept in a 37C water bath for four hours and mixed using a vortex for 2 min every 30 min. All PEG and
DEX solutions were filtered through syringe filters of 0.2 μm pore size to remove impurities. Resulting polymer solutions were stored at 4C.
2.1.2 Construction of Binodal Curves
27
Binodal curves for systems A, B, C, and D were constructed using a titration method [174]. Various ATPS were prepared in a complete growth medium in 1.5 mL microcentrifuge tubes from stock PEG and DEX solutions. Medium was added to each ATPS in 5 µl increments until the interface between the top and bottom phases disappeared. Concentrations of the polymers prior to formation of one phase, known as a node, were determined, and used to construct a binodal curve.
2.1.3 Cell Culture
BT474 breast cancer cells were obtained from ATCC and cultured in RPMI
1640 medium (Sigma) supplemented with 10% fetal bovine serum (FBS) (Sigma),
1% of 1 mM sodium pyruvate solution (Life Technologies), 1% of 0.1 mM nonessential amino acids solution (Life Technologies), 1% of 10 mm Hepes buffer solution (Life Technologies), and 1% of antibiotic solution (Life Technologies).
Cells were cultured in a humidified incubator at 37C and 5% CO2 in a T75 flask
(Thermo Fisher Scientific). BT474 cells grew in multilayer patches. Cells were rinsed with phosphate buffered saline (PBS) (Sigma) and detached using a 0.25% trypsin solution (Life Technologies) for cell seeding and passaging. Cells were sub- cultured at a ratio of 1:3.
2.1.4 Preparation of Collagen Solution
A stock solution of type I rat tail collagen (Corning) with a concentration of
8.56 mg/ml dissolved in 0.02N acetic acid was diluted to desired concentrations using the manufacturer’s protocol. For example, 1 mL of 4 mg/ml collagen solution 28 was prepared by mixing 100 µl from a 10X DMEM medium, 422 µl sterile distill water, 467 µl collagen stock solution, and 11 µl from a 1N NaOH solution. All the reagents were kept on ice during collagen preparation to maintain the temperature at 4C and prevent premature gelation of collagen. The pH of the solution was measured using a pH meter (Mettler Toledo) and maintained at 7.5. Prepared collagen solutions were stored at 4C for a maximum of 1 h before use.
2.1.5 Partition of Collagen in ATPS
Collagen partition experiments were performed with systems A-D in serum- free RPMI. From each system, three combinations were used: 15% PEG-21%
DEX, 18% PEG-24% DEX, and 24% PEG-32% DEX. Equal volumes (200 µl) of a
PEG solution, a DEX solution, and a 2 mg/ml type I collagen solution were mixed in a 1.5 mL microcentrifuge tube. The mixture was equilibrated at 4C for 60 min to allow collagen to partition between the two phases and a clear interface form.
Because collagen partition experiments were conducted with equal volumes of
PEG, DEX, and collagen solutions, the resulting concentrations of PEG and DEX in each system in the partition assay reduced to 5% PEG-7% DEX, 6% PEG-8%
DEX, and 8% PEG-10.6% DEX. Four replicates were used for each system. From each tube, the top phase solution was pipetted out first followed by the bottom phase solution. Samples were stored in separate microcentrifuge tubes.
2.1.6 Hydroxyproline Assay
29
Collagen concentration in the bottom phase of each system was quantified using a hydroxyproline assay (Sigma). Briefly, equal volumes of the sample and concentrated hydrochloric acid (~12 M HCl) were mixed in a Teflon-capped glass vial (Taylor Scientific). Next, the solution was hydrolyzed at 110C in a hot air oven
(Binder) for 16 h. Then, the vial was cooled down to room temperature and the hydrolyzed solution was transferred into a 1.5 mL centrifuge tube. The tube was centrifuged for 10 min at 180 rcf for 10 min. Next, 10 µl of the supernatant was transferred into a flat-bottom 96-well plate. The plate was incubated in a 60C oven with the lid on for 2 h for complete drying of the sample. Each sample was spiked with 0.4 µg of hydroxyproline standard to remove absorbance interference from endogenous compounds. Hydroxyproline standards were also run simultaneously to obtain a standard curve. To each well, 6 µl of chloramine T concentrate and 94
µl of an oxidation buffer was added. The plate was incubated at room temperature for 5 min. This was followed by addition of 50 µl 4-(Dimethylamino) benzaldehyde concentrate and 50 µl perchloric acid. The plate was incubated at 60C for 90 min and absorbance was measured at 560 nm using a plate reader (Synergy H1M)
(BioTek Instruments). The hydroxyproline standard curve was used to resolve the hydroxyproline amino acid content in the sample. The collagen content was approximated by multiplying the resulting value by a factor of 7.69 [175].
2.1.7 Cancer Cell Spheroid Formation
30
BT474 cancer cell spheroids were formed using our ATPS technology [128].
PEG 35 kDa and DEX 500 kDa were used to form the spheroids. Aqueous PEG phase solution of 6.6% (w/v) and DEX phase solution of 3.2% (w/v) were prepared separately in a complete growth medium [128]. BT474 cells were mixed thoroughly with the DEX phase solution to form a cell suspension with a density of 50×103 cells/μl. Next, 30 μl of the PEG solution was loaded into a round-bottom, ultralow attachment 384-well plate (Corning). A 0.3 μl drop of the DEX phase containing
15×103 cells was dispensed into each well using a robotic liquid handler (SRT
Bravo) (Agilent Technologies). The plate was incubated at 37C for 24 h to allow formation of a spheroid in each DEX phase drop within each microwell. Phase contrast images of spheroids were captured using an inverted fluorescence microscope (Axio Observer A1) (Zeiss).
2.1.8 Formation of Collagen-Embedded Cancer Cell Spheroids
An aqueous phase solution of 14% (w/v) DEX 500 kDa was prepared in a complete growth medium and mixed with an equal volume of 4 mg/ml collagen solution to obtain a solution of 7% (w/v) DEX and 2 mg/ml collagen. After BT474 spheroids formed, 10 μl of the collagen-DEX solution was dispensed into each well containing a spheroid submerged in the PEG solution. These concentrations of
PEG, DEX, and collagen were selected to replicate partition of 2 mg/ml collagen in an ATPS of 5% (w/v) PEG-7% (w/v) DEX system. The 384-well plate containing spheroids was maintained at 4C for 30 min before dispensing the collagen-DEX
31 solution. Again, the robotic liquid handler was used for uniform dispensing of the solution. The 384-well plate was kept on an ice tray for 60 min to allow collagen partitioning take pace and another 30 min in room temperature. Then, the plate was incubated at 37C to allow the collagen to gel. The 384-well plate was not transferred directly from 4C to 37C incubator to prevent potential thermal shock to cells.
2.1.9 Statistical Analysis
Data from the experiments were expressed as mean ± standard error. Two- way ANOVA with Bonferroni post hoc tests (MINITAB) were used to compare means among groups. Each group had at least n=4 and p < 0.05 defined the level of statistical significance.
2.2 Results
Below the results of characterization of different ATPS systems and utilization of one ATPS system to encapsulate cancer spheroid in a partitioned collagen are presented.
2.2.1 Characterization of ATPS
We used a titration method to construct binodal curves for four systems A-
D formed with different molecular weights of PEG and DEX [176]. Each curve represents critical concentrations of phase-forming polymers above which two distinct aqueous phases formed (Figure 2.1). Construction of binodal curves was 32 necessary to determine the working concentrations of polymers that give two- phase systems. The stock concentrations of PEG and DEX (15% PEG-21% DEX,
18% PEG-24% DEX, and 24% PEG-32% DEX) were selected because two-phase solutions made at these concentrations contain equal volumes of PEG-rich top phase and DEX-rich bottom phase, making it convenient to measure the volume of each phase during partition experiments. This also allowed visual comparison of partition of collagen in the bottom phase of two-phase systems made with the three sets of concentration pairs in each system A-D. As expected, the binodal curve was more asymmetric when the difference in molecular weights of PEG and
DEX polymers increased [176]. System C had the most symmetric binodal curve, whereas system B had the most asymmetric binodal curve (Figure 2.1).
2.2.2 Standard Curves of Collagen Concentration Using Hydroxyproline Assay
We quantified collagen contents in samples using a commercially available hydroxyproline assay kit. To assess the accuracy of the assay, we prepared known concentrations (1, 2, 3, 4, 6, and 8 mg/mL) of collagen standards in serum free
RPMI medium to avoid interference of serum proteins with the hydroxyproline amino acids present in the collagen standards. We acid-hydrolyzed collagen standards for 16 h and determined the protein concentration. The result in Figure
2.2a shows a strong correlation between the actual collagen concentrations used in the experiment and the measured collagen concentrations, indicating that this assay can precisely predict collagen concentration in sample solutions.
Importantly, we generated hydroxyproline standard curves using distilled water 33 and PEG and DEX solutions, and showed that polymers, at least in the concentrations used, do not cause interference in the absorbance signal from the samples when quantifying collagen concentration in an aqueous polymeric solution (Figure 2.2b).
2.2.3 Collagen Partition in ATPS
We performed collagen partition experiments with systems A-D. We mixed equal volumes of PEG solution, DEX solution, and collagen solution in a 1.5 microcentrifuge tube, equilibrated the mixture at 4C, and transferred top and bottom phase into separate vials (Figure 2.3). From each system, we selected three pairs of concentrations of aqueous phases: 15% PEG-21% DEX, 18% PEG-
24% DEX, and 24% PEG-32% DEX. The resulting polymer concentrations in the collagen partition assay were 5% PEG-7% DEX, 6% PEG-8% DEX, and 8% PEG-
10.2% DEX, respectively (Figure 2.1). These pairs located above the binodal curve and resulted in two-phase systems which we visually confirmed by observing a clear interface. We defined a partition coefficient (Kc) as the ratio of collagen concentration in bottom phase and the total collagen concentration used in each assay.
collagen concentration in bottom phase Kc = × 100% total collagen concentration
Figure 2.4 shows an image from a collagen partition assay using the 5%
PEG-7% DEX pair of system A. A distinct interface between the PEG-rich top phase and DEX-rich bottom phase is clear. This two-phase system gave a collagen
34 partition coefficient of 61±3%. Figure 2.5 shows partition coefficients of collagen in systems A-D, each made with the 5% PEG-7% DEX pair. Collagen partition coefficient was the highest in system A (61±3%). When the molecular weight of
DEX increased to 500 kDa but the molecular weight of PEG was kept constant, the partition coefficient significantly decreased to 33±4% in system B. Increasing the molecular weight of PEG from 8 kDa in system B to 35 kDa in system D but keeping the molecular weight of DEX constant significantly increased the partition coefficient to 58±2%. With systems A-D, we obtained similar results using 6%
PEG-8% DEX and 8% PEG-10.2% DEX pairs of concentrations (Figure 2.6).
2.2.4 Collagen Partitioning for Tumor Tissue Engineering
A previous study showed that collagen primarily partitions to the interface of ATPS [173]. This property was used to generate low-volume collagen microdrops (<10 µL) that mimicked matrix contraction in tissue environments.
Here, we demonstrated the utility of partitioning collagen to the DEX phase of
ATPS to create a 3D tumor microtissue. In our study, we quantitatively showed improved partition of collagen to DEX phase of several ATPS. We selected system
D with 5% PEG-7% DEX because our preliminary experiments showed that the
PEG 35 kDa –DEX 500 kDa ATPS gives consistently-sized and compact spheroids over a wide range of polymer concentrations compared to other ATPS used [177].
From this ATPS, we used the 5% PEG-7% DEX pair because it gave a similar collagen partition coefficient to those from the 6% PEG-8% DEX and 8% PEG-
35
10.2% DEX pairs (Figures 2.5 and 2.6) but at a reduced concentration of the polymers that is preferable for cell-based applications with ATPS [128,136]. After a spheroid formed in each well of a 384-well plate, we dispensed 10 µl of the DEX solution containing collagen to the wells containing spheroids on the well-bottom submerged in the PEG phase solution (Figure 2.7a). Because the DEX phase is denser than the PEG phase, it sank to the bottom of the wells and due to the propensity of collagen towards the DEX phase, it remained in the DEX phase during incubation. We confirmed confinement of collagen to the DEX phase by dispensing equal amounts of the PEG and DEX phases and collagen in a PCR tube (Figure 2.7b). Incubating the plate at 37C led to the gelation of collagen embedding of the spheroid (Figure 2.7c). We visually confirmed that the spheroid was embedded in collagen by performing this assay in a PCR tube (Figure 2.7d).
2.3 Discussion
We developed a novel method of formation of 3D tumor model using cell and protein partitioning in ATPS and in two convenient steps by first generating a spheroid and then encapsulating it in a collagen matrix. We utilized aqueous PEG and DEX polymeric system to partition collagen solution to bottom DEX phase.
Results suggest that collagen partition in ATPS is highly sensitive to molecular weights of phase polymers; that is, reducing the molecular weight of a polymer increases the propensity of collagen to partition to the aqueous phase of that polymer. This finding is consistent with a previous report that proteins in non-ionic
36
ATPS attract to the aqueous phase with smaller polymer molecules if all other conditions such as polymer concentration, temperature, and salt concentrations are kept constant [178]. Another interesting finding emerged from our results. That is, increasing PEG molecular weight from 8 kDa in system A to 35 kDa in system
C but keeping DEX molecular weight fixed at 40 kDa still increased the partition coefficient by 5%. However, this increase was significantly smaller than that from system B to system D when DEX molecular weight was 500 kDa. This suggests that in addition to the effect of changing the molecular weight of one of the phase polymers on collagen partition, greater differences between molecular weights of the two-phase polymers is required to make a significant impact on uneven partition of the protein between the aqueous phases of the polymers. Finally, comparing results among the three pairs of PEG and DEX in a specific system, i.e., 5% PEG-7% DEX, 6% PEG-8% DEX, and 8% PEG-10.2% DEX, showed no statistical difference in the collagen partition coefficient (Figures 2.5 and 2.6).
Collagen embedding of spheroids has been previously done using different approaches. For example, after forming spheroids using a rocking method, spheroids were transferred into a well plate and then overlaid with collagen [179].
Liver cell spheroids were formed using a hanging drop method and multiple spheroids were transferred to a well plate containing a collagen solution to produce
3D culture in a collagen hydrogel [180]. These methods of forming spheroids have several limitations such as evaporation of media from the hanging drops and difficulty in handling the hanging drop culture plates, inconsistently-sized spheroids
37 made with the plate rocking technique, etc. [129]. More importantly, spheroids had to manually be transferred to another plate and the media had to be removed before the collagen solution was dispensed onto the spheroids. This approach is labor-intensive and risks losing the spheroid during aspiration of medium. Unlike these methods, our method eliminates the tedious process of transferring spheroids to a second plate to embed them in collagen. We use the same ATPS formulation both to prepare spheroids and to partition collagen to embed the spheroids. The entire process is done in two pipetting steps: First, a DEX phase drop containing cancer cells is dispensed into the PEG phase to form a spheroid in the drop phase. Then, a collagen-containing DEX drop is dispensed to merge with the spheroid-containing drop and form a hydrogel that entraps the spheroid
(Figure 2.7a). This approach significantly simplifies the preparation of microtissues. Additionally, because partitioning of proteins in ATPS is independent of polymer concentrations [181], this approach can conveniently produce collagen gels of desired concentrations to reproduce mechanical properties of tumors in vivo [182], matrix stiffness and porosity [183], and collagen permeability [184]. Our quantitative results of partition of collagen in ATPS (Figures 2.5 and 2.6) is key to facilitate this approach. This also allows us to produce hydrogels of different sizes and stiffness values simply by changing the volume of the DEX phase drops containing desired concentrations of collagen. This simplified approach is especially a major advantage for high throughput applications such as cancer drug screening [185]. Encapsulating spheroids using collagen partitioning in ATPS is a
38 novel technique to develop an in vitro 3D tumor model. More complex tumor models can also be conveniently developed by including other cellular components of the TME such as fibroblasts and immune cells [20,133].
2.4 Summary
We used a quantitative approach to establish that collagen partition in ATPS is sensitive to polymer molecular weight. Using this property, we improved partition of collagen to the DEX phase of a PEG-DEX ATPS and employed this approach to conveniently develop 3D breast tumor models. This new technique will enable future studies to investigate the impact of components of the TME on different functions of cancer cells.
39
Figure 2.1: Binodal curves of four systems made with different molecular weight PEG and DEX: (a) PEG 8 kDa-DEX 40 kDa, (b) PEG 8 kDa-DEX 500 kDa, (c) PEG 35 kDa-DEX 40 kDa, and (d) PEG 35 kDa-DEX 500 kDa. Colored points show the resulting concentrations of PEG and DEX in the partition assay; square: 5%PEG-7%DEX, triangle: 6%PEG-8%DEX, and circle: 8%PEG-10.2%DEX. All concentrations of aqueous PEG and DEX solutions were calculated in %(w/v).
40
Figure 2.2: Standard curves for (a) hydroxyproline assay and (b) for hydroxyproline standards prepared in distilled water, a 10%(w/v) DEX 500 kDa solution, and a 10%(w/v) PEG 35 kDa solution. All samples were hydrolyzed, and absorbance was measured following the manufacturer's protocol. Statistical method: two-way ANOVA. NS: statistically non-significant. Error bars are standard error of the mean. n=5.
41
PEG DEX Collagen Top Bottom phase phase solution phase phase
Mix Equilibrate (4oC) Separate
Figure 2.3: Schematic of collagen partition in aqueous two-phase systems.
42
PEG-rich top phase
interface
DEX-rich bottom phase
Figure 2.4: Collagen partition assay containing equal volumes of 15% PEG 35 kDa, 21% DEX 500 kDa, and 2 mg/ml collagen resulting in formation of 5% PEG- 7% DEX ATPS. All concentrations of aqueous PEG and DEX solutions were calculated in %(w/v).
43
Figure 2.5: Collagen partition coefficient (Kc) in four different systems using the 5% PEG-7% DEX concentration pair. *p<0.05, n=4. All concentrations of aqueous PEG and DEX solutions were calculated in %(w/v).
44
Figure 2.6: Collagen partition coefficient (Kc) in four different systems in 5% PEG- 7% DEX, 6% PEG-8% DEX, and 8% PEG-10.2% DEX concentration pairs. *p<0.05, n=4. All concentrations of aqueous PEG and DEX solutions were calculated in %(w/v).
45
Figure 2.7: (a) Schematic of embedding a spheroid in partitioned collagen. (b) From a 14% DEX 500 kDa solution mixed with 5 µl of 4 mg/ml collagen solution, a 10 µl volume is dispensed in a tube containing 30 µl of 6.6% PEG 35 kDa. Collagen gels in the bottom phase. (c,d) Top view of a BT474 cancer cell spheroid embedded in a collagen gel partitioned to the DEX phase in a microwell. (d) Side view of a BT474 spheroid in a collagen gel partitioned to the DEX phase in a PCR tube. All concentrations of aqueous PEG and DEX solutions were calculated in %(w/v).
46 CHAPTER III
ORGANOTYPIC BREAST TUMOR MODEL ELUCIDATES DYNAMIC REMODELING OF TUMOR MICROENVIRONMENT
Portions of this chapter are reused from:
S. Singh, L.A. Ray, P.S. Thakuri, S. Tran, M.C. Konopka, G.D. Luker, H. Tavana, Organotypic breast tumor model elucidates dynamic remodeling of tumor microenvironment, Biomaterials. 238 (2020). © 2020 Elsevier Ltd. https://pubmed.ncbi.nlm.nih.gov/32062146/
The tumor microenvironment (TME) is a complex tissue containing cancer cells and various stromal cells such as fibroblasts, immune cells, and endothelial cells embedded in an extracellular matrix (ECM) network [20]. The stromal cells and the ECM are collectively referred to as the tumor stroma and interactions among cancer cells and different components of the stroma are known as tumor- stromal interactions. Among different stromal cells, fibroblasts play pivotal roles in initiation and progression of epithelial tumors [186,187]. During tumor progression, cancer cells secrete signaling molecules to recruit resident fibroblasts of the tissue and activate them. These cells, known as cancer-associated fibroblasts (CAFs), are the most abundant stromal cells in solid tumors and associate with cancer cells at all stages of tumor progression [186,188]. Histopathological analysis of human
47 tumors shows that advanced tumors with poor prognosis are abundant in CAFs
[44]. In breast cancer, CAFs and their ECM cause a tumor to be palpable [189].
When epithelial cancer cells breach the basement membrane and invade the ECM, they come in direct contact with CAFs [190]. Additionally, CAFs secrete various pro-invasive signaling molecules into the tumor microenvironment and deposit and remodel the ECM [191]. Physical and biochemical interactions with CAFs and their associated ECM promote proliferation of cancer cells and alter their morphology to an invasive mesenchymal-like shape [192]. Therefore, CAFs are a critical component of solid tumors and a key regulator of tumor-stromal interactions.
Considering that traditional treatments of solid tumors tailored toward eliminating cancer cells are largely ineffective, targeting of tumor-stromal interactions is being pursued to develop new therapies.
Cell cultures and animal models are typically used to study tumor-stromal interactions. Culture of cancer cells and fibroblasts in an intermixed monolayer or in separated sheets provides a convenient method to study interactions between the two cell types. However, monolayer cultures do not reproduce complexities of tumors that are three-dimensional (3D), including oxygen and nutrients distribution profiles, cell-cell and cell-ECM interactions, diffusive barrier to drug delivery, etc.
[193,194]. Mouse models provide physiological systems for cancer research but lack human tumor stroma and need complex biological assays and sophisticated imaging to study effects of stroma on tumor cells [195]. To address the technological need for physiologic human tumor models, 3D cultures of cancer
48
cells as spheroids have been developed. Spheroids are compact clusters of cells that reproduce properties of solid tumors such as diffusive transport of oxygen and nutrients, hypoxia and necrosis, drug delivery barriers, and deposition and degradation of ECM proteins [117,196,197]. Co-culture spheroids of cancer cells with various stromal cells have been used to study tumor-stromal interactions that promote proliferation and drug resistance of breast cancer cells [133], and their epithelial-mesenchymal transition (EMT) [122] . Despite their benefits, co-culture spheroids do not mimic the architecture of native solid tumors in terms of spatial distribution of cancer and stromal cells and cell-ECM interactions. Matrigel has been used to develop spheroid cultures to study cell-cell and cell-ECM interactions and ECM invasion of cancer cells [198,199]. However, Matrigel is an animal- derived protein cocktail and not amenable to adjusting its composition and physicochemical properties [150]. To gain control over spatial distribution of cells and ECM properties, microfluidics technology has been widely used to develop 3D tumor models. Recently, microfluidic co-culture systems containing hydrogel- based 3D matrices were used to study tumor-stromal interactions, cancer cell invasion of ECM, and anti-metastatic drug testing [157,200–202]. Major drawbacks of microfluidic cell culture devices are their complexity and need for user expertise to develop and maintain them especially for long-term cultures, incompatibility with industrial-scale high throughput drug screening applications, and adsorption of drugs and proteins on the surface of polydimethylsiloxane (PDMS) microfluidic devices [203,204].
49
To overcome these limitations of existing cultures, we developed a 3D organotypic breast tumor model that resembles the architecture of avascular solid tumors and consists of three major components of the TME: a mass of cancer cells, dispersed fibroblasts, and ECM. We also adapted this technology to a microwell plate format to enable convenient, reproducible formation of cultures, high throughput drug screening, and adaption by other users. To establish the validity of our model, we used triple negative breast cancer (TNBC) with CXCL12
– CXCR4 signaling as a disease model. CXCL12, or stromal-derived factor-1α
(SDF-1α), is a major paracrine signaling molecule produced by CAFs in the tumor microenvironment. CXCL12 signals through its cognate CXCR4 receptor often overexpressed on the surface of TNBC cells [205,206]. Therefore, we use engineered CXCL12-secreting fibroblasts and CXCR4-expressing TNBC cells in our microtissues to recapitulate paracrine stromal-cancer cells signaling. Using different molecular analyses and imaging techniques, we demonstrated that
CXCL12-secreting fibroblasts significantly contract the ECM. Additionally, signaling between the fibroblasts and TNBC cells promoted ECM invasion of cancer cells by activating oncogenic mitogen-activated protein kinase (MAPK) pathway. We also demonstrated that blocking tumor-stromal interactions inhibits cancer cell invasiveness as a potential treatment strategy. This organotypic model is a convenient-to-adapt tool broadly useful to study different cancers. It offers straightforward implementation to facilitate dissemination to academic and industrial settings, allows adjusting added components including single to multiple
50 cell types and ECM, and its compatibility with microwell plates uniquely enables high throughput drug testing against tumor-stromal interactions to identify novel therapeutics.
3.1 Materials and Methods
The methods for spheroid formation, immunofluorescence of 3D cultures of dispersed fibroblasts, microtissue formation, and ECM invasion are described below.
3.1.1 Cell Culture
Two different fibroblast cells were used: normal human mammary fibroblasts (gift of Dr. Daniel Hayes, University of Michigan) stably transduced with mCherry protein (labeled HMFs) and human mammary fibroblasts stably transduced to secrete 0.75 ng/ml/h of CXCL12α fused with mCherry (labeled
CXCL12+HMFs). MDA-MB-231 TNBC cells (purchased from the ATCC and labeled TNBC) overexpressing CXCR4 receptors (labeled CXCR4+TNBC) were transduced to stably express GFP [160]. These cells were generated in Dr. Luker’s laboratory in University of Michigan. The authenticity of cells after stable transduction was verified by short tandem repeat profiling. Cells were tested for mycoplasma and maintained in Plasmocin prophylactic (Invivogen) according to the manufacturer’s directions. Cells were cultured in Dulbecco’s Modified Eagle
Medium (DMEM, Sigma) supplemented with 10% fetal bovine serum (FBS,
Sigma), 1% glutamine (Life Technologies), and 1% antibiotic (Life Technologies). 51
Cells were cultured in a humidified incubator at 37C and 5% CO2 in a T75 flask
(Thermo Fisher Scientific). When the cells were about 80% confluent, they were rinsed with phosphate buffered saline (PBS, Sigma), dislodged using 0.25% trypsin (Life Technologies), and sub-cultured.
3.1.2 Culture of Fibroblasts in Collagen
Fibroblast cells were harvested from a confluent monolayer and mixed with an ice-cold collagen solution. Collagen solutions of desired concentrations were prepared from stock solutions of type I rat tail collagen (Corning) using the manufacturer’s protocol [207]. All the reagents were kept on ice during collagen preparation to prevent premature gelation of collagen. The pH of solutions was measured with a pH meter (Mettler Toledo) and maintained at 7.5 by adding 1 N
NaOH or 1 N HCL dropwise. Of the stock collagen solution, 1 mg/ml and 4 mg/ml solutions were prepared and stored at 4C for up to 1 h before use. The 1 mg/ml and 4 mg/ml collagen gels are referred to as softer and stiffer gels, respectively, for convenience. Different densities (number of cells per well) of HMFs and
CXCL12+HMFs were used to study collagen contractility. For example, to form stiffer collagen gels with 1.5 × 104 cells in each well of a 384-well plate, 7.5 × 105 fibroblasts were mixed with 1 ml of 4 mg/ml collagen solution, and 20 µl of the resulting suspension solution was dispensed into each well. Incubation at 37C for
30 min resulted in hydrogel formation. Cultures were maintained in DMEM supplemented with only 1% FBS. Phase images were captured using an inverted
52 fluorescence microscope (Axio Observer A1, Zeiss). Contraction of collagen gels was quantified as a matrix contractility = (1 – ratio of projected area of a culture before and after contraction). Prior to this experiment, nanoindentation of collagen gels prepared from various concentration of 1.0, 2.5, 4.0, 5.5, 7.0, and 8.5 mg/ml was conducted using an atomic force microscope (AFM) (Digital Instruments).
Plasma cleaning and UV radiation were performed on indentation probe and tip- holder to remove organic contaminants before each experiment. PBS buffer was used as a rehydrating medium for gels during indentation. We used 300 nm radius silicon-oxide probe with cantilever spring constant of 0.06 N/m. Velocity of extension and recession of cantilever during indentation was controlled at 500 nm/s, slow enough to avoid hydrodynamic force from flow of buffer solution. A
Poisson’s ratio of 0.5 was used. At least 30 indentations were made on each sample to generate force versus indentation curves that were analyzed using Hertz model to generate elastic modulus values for collagen gels of various protein concentrations.
3.1.3 Spheroid Formation
CXCR4+TNBC spheroids were formed using our established protocol with an aqueous two-phase system [136]. Polyethylene glycol (PEG) (Mw: 35 kDa,
Sigma) and dextran (DEX) (Mw: 500 kDa, Pharmacosmos) were used to form 5%
(w/v) aqueous PEG phase and 6.4% (w/v) aqueous DEX phase solutions in a complete growth medium [128,129]. The PEG phase solution was supplemented with 0.24% of methylcellulose powder. CXCR4+TNBC cells were mixed thoroughly 53 with the DEX phase solution to form a cell suspension with a density of 2.5×104 cells/µl. Next, 30 µl of the PEG solution was loaded into a round-bottom, ultralow attachment 384-well plate (Corning) labeled as the destination plate. A robotic liquid handler (SRT Bravo, Agilent Technologies) was used to dispense a 0.3 µl drop of the DEX phase containing 7.5×104 cells into each well of the destination plate [208]. This process was done column-by-column to reduce the need for large volumes of high-density cell suspensions. The plate was incubated at 37C for 24 h to allow formation of a spheroid in the DEX phase drop within each microwell.
3.1.4 Immunofluorescence of 3D Cultures of Dispersed Fibroblasts
After 5 days of incubating collagen gels containing only fibroblasts (1.5 x104 cells), cultures were transferred from the microwells into a Petri dish using a micro- spatula (Fine Science Tools). The gels were fixed using 4% paraformaldehyde
(PFA) in PBS for 20 min at room temperature and washed three times with PBS, each time for 10 min. The gels were blocked for 1 h at room temperature with 5% goat serum. Cells were permeabilized using 0.3% Triton X-100 in PBS. Samples were incubated overnight with a mouse anti-β-actin antibody (1:200 dilution, ab6276, Abcam) prepared in 1% BSA and 0.3% Triton X-100 in PBS. Gels were washed and then incubated with a goat anti-mouse FITC-conjugated secondary antibody (Jackson ImmunoResearch) for 1 h followed by nuclear staining with
Hoechst 33342 (Thermo Fisher Scientific) and imaged at 60x with a confocal microscope (Nikon A1). FITC was excited by a 488 nm laser and the emission
54 isolated with a 500-550 nm filter, while Hoechst was excited by 405 nm and isolated with a 425-475 nm emission filter.
3.1.5 Metabolic Activity of Fibroblast Cells in 3D Cultures
HMFs and CXCL12+HMFs were added to separate collagen solutions in a
384-welll plate at a density of 1.5 x104 cells per well. The collagen solution was gelled and medium with 1% FBS was added. The growth of the fibroblasts in collagen was evaluated for 2 weeks using a standard PrestoBlue metabolic activity assay (Invitrogen). After 15 min of incubation, the fluorescent signal was measured with a standard plate reader (Synergy H1M, BioTek Instruments) at 560 nm excitation and 590 nm emission wavelengths. Collagen gels without fibroblasts were used as a negative control.
3.1.6 Conditioned Medium Cultures
HMFs cultures in 3D collagen gels were treated with conditioned medium collected from a confluent monolayer of CXCL12+HMFs. CXCL12+HMFs cells were initially seeded in a flask containing 1% FBS growth media. Conditioned medium was added to the HMFs culture every other day by replacing 60 µl of existing medium. Contraction of ECM was quantified using phase contrast images.
3.1.7 Wound Healing Assay
Both HMFs and CXCL12+HMFs were seeded as a monolayer culture in 35 mm Petri dishes (Corning) and cultured until confluent. Using a pipette tip, a 55 straight scratch was made in the cell layer. Cells were imaged for 48 h and the
A2 migration of cells in the cultures was quantified as gap closure= 1 − , where A1 A1 is the initial gap area and A2 is the final gap area.
3.1.8 Organotypic Microtissue Formation, ECM Invasion, and AMD3100 Treatment
After formation of compact spheroids, medium was robotically removed from the wells. The protocol was optimized to ensure that spheroids remained in the wells during aspiration. This was done by adjusting the height of the pipette tip from the bottom of each well to 0.2 mm, maintaining a low aspiration rate of 0.6
µl/sec, and offsetting the position of the pipette tip 30% from the center of the well where spheroids were located. Cold collagen solution containing 1.5×104 fibroblasts (HMFs or CXCL12+HMFs) was dispensed into each well containing a spheroid. The well plate was incubated at 37C for 30 min to form a collagen gel and then 60 µl of medium containing only 1% FBS was added (Figure 3.1a). For convenience, microtissues containing HMFs or CXCL12+HMFs are labelled
+HMFs and +CXCL12+HMFs, respectively. As a negative control, microtissues without fibroblasts were generated by adding the collagen solution to wells containing a cancer cell spheroid. Medium was refreshed every 72 hrs. Both aspirating and dispensing of collagen solution and medium were carried out robotically for accuracy and speed of sample preparation. Confocal images were taken with a 10X objective on days 1, 3, and 5 to assess matrix invasion of cancer cells. A 488 nm laser was used to excite the fluorophore, while a 500-550 nm filter 56 was used to capture GFP fluorescence from CXCR4+TNBC cells. Z-stacks were constructed from images of samples acquired with a z-spacing of 20 μm. NIS software was used for image acquisition and Fiji (ImageJ, NIH) was used for analysis and 3D reconstruction. CXCR4+TNBC cell invasion of the ECM was quantified by analyzing images from both z-projected images and from the individual images of each z-stack.
ECM invasion of cells was analyzed in two different ways. First, all the confocal images from each sample were collapsed to form a z-projected image.
This image was opened in ImageJ and the scale was removed to set the scale to pixels and measure area in pixels. Then, color thresholding was used to select the area covered by the cells, and pixel areas were measured to quantify invasion area. Because this commonly used method masks vertical displacement of cells, the vertical scattering of cancer cells was also analyzed. Briefly, each image from a stack of the confocal images was analyzed to measure the pixel area occupied by the CXCR4+TNBC cells. The area of pixels (labeled as invasion area) was then plotted against the z-distance of the images from each stack to obtain a histogram.
Each microtissue resulted in one histogram and each data point in the histogram represented pixels area from one image. Area under the curve (AUC) was measured from each histogram in MATLAB (MathWorks). AUC values were averaged to express as 3D invasion cell area.
Because dispersed fibroblasts contracted the collagen matrix, the contractility of the matrix in each microtissue containing a cancer cell spheroid and
57 fibroblasts was expressed between 0 – 1, with 1 being the highest matrix contraction. Then, ECM invasion of the CXCR4+TNBC cells was normalized with contractility and expressed as
normalized invasion cell area= 1 * invasion cell area (1− 푚푎푡푟𝑖푥 푐표푛푡푟푎푐푡𝑖푙𝑖푡푦 )
To block signaling from CXCL12+HMFs to CXCR4+TNBC cells, microtissues were treated with 1 µM of AMD3100. Medium change was done every other day using 40 µl of fresh medium containing AMD3100. ECM invasion of
CXCR4+TNBC cells was imaged using a confocal microscope and analyzed as described above.
3.1.9 Western blotting
Microtissues were treated with collagenase I (Sigma) for about 10 min with added mechanical agitation by pipetting to dissociate and dissolve the matrix.
Collagenase activity was neutralized by adding triple amounts of complete growth medium using the manufacturer’s protocol. The resulting suspension was centrifuged to obtain a cell pellet. The cells were washed with PBS and lysed with
500 µl of complete RIPA buffer containing 1% protease inhibitors and 1% phosphatase inhibitors. The suspension was further sonicated twice at a 20% amplitude for 5 sec (Vibra-Cell, Sonics). A BCA protein quantification assay (Life
Technologies) was used to measure total protein concentration from the samples, of which 25 µl was loaded onto a 4-15% gel (Bio-rad) for electrophoresis. The gel was transferred onto a nitrocellulose membrane by electroblotting, blocked in 5%
58 non-fat dried milk prepared in wash buffer for 1 h, and incubated overnight with a primary antibody. Primary antibodies used were rabbit anti-phospho-Myosin Light
Chain 2 (Thr18/Ser19) antibody (Cat. No. 3674), rabbit anti-Myosin Light Chain 2 antibody (Cat. No. 3672), rabbit anti-phospho-PI3K antibody (Cat. No. 4228), rabbit anti-phospho-p44/42 MAPK (p-Erk1/2) antibody (Cat. No. 9101), rabbit anti- p44/42 MAPK (Erk1/2) antibody (Cat. No. 9102), rabbit anti-phospho-Akt (Ser473) antibody (Cat. No. 4060), rabbit anti-Akt (Cat No. 9272), and GAPDH antibody
(Cat. No. 5174), all purchased from Cell Signaling Technology. The membranes were washed and then incubated with anti-rabbit horseradish peroxidase (HRP)- conjugated secondary antibody (Cell Signaling Technology, Cat. No. 7074) for 1 h. Detection was carried out using ECL chemiluminescence detection kit (GE
Healthcare) with FluorChem E imaging system (ProteinSimple).
3.1.10 Statistical Analysis
Data were first checked for normality using Anderson-Darling method in
MINITAB. For normally distributed data, one-way ANOVA with Tukey’s pairwise comparisons were used to compare means among 3 or more samples. Two- tailed, unpaired t-test was used to compare two experimental groups. Normally distributed data from the experiments were expressed as mean ± standard error.
Non-Gaussian distributed data were analyzed using two-tailed Mann Whitney test. Values of p < 0.05 were used to define statistical significance.
59
3.2 Results
Below the results of validating the organotypic cultures biologically using a specific CXCL12-CXCR4 pathway are presented. This includes generation of 3D organotypic cultures, mechanisms of ECM contraction by fibroblasts, role of fibroblasts in matrix invasion of cancer cells, and inhibition of ECM invasion of
CXCR4+ TNBC cells.
3.2.1 Fibroblasts Contract the ECM
To reproduce solid breast tumors, we generated organotypic tumor models containing a CXCR4+TNBC cell mass within a collagen matrix with dispersed fibroblasts (HMFs or CXCL12+HMFs). After forming a compact CXCR4+TNBC spheroid using our ATPS technology, we aspirated the medium robotically, dispensed collagen solution containing fibroblasts to embed the spheroid in each well and incubated the plate to allow collagen to gel (Figure 3.1a). Prior to this, we also performed nanoindentation of collagen gels to determine the elastic modulus value of the collagen gel to match that of breast tumor tissues (Figure 3.2a,b).
Figure 3.1b shows a 3D reconstructed volume view of the microtissue where a spheroid of CXCR4+TNBC cells (MDA-MB-231; blue) is surrounded by fibroblasts
(green) dispersed in the collagen gel.
We observed that the cultures detached from walls of microwells and shrank over time. To understand this phenomenon, we created microtissues of collagen- embedded CXCR4+TNBC spheroids only or collagen-embedded dispersed fibroblasts only. Cultures containing only CXCR4+TNBC spheroids in collagen did 60 not shrink the gel even after two weeks of culture, whereas both HMFs and
CXCL12+HMFs contracted the collagen gel time-dependently. To quantitatively study contraction of the ECM by fibroblasts, we used known numbers of fibroblast cells in collagen gels of two different protein concentrations. We first examined the spatial distribution of fibroblasts in collagen and consistency of preparation.
Fibroblasts were uniformity distributed in the 3D space and the number of cells included closely matched the number of cells quantified from the confocal images
(Figure 3.3a,b). Next, we captured phase images of the cultures and quantified
ECM contraction over time. This study considered effects of collagen gel elastic modulus, fibroblast cell density, and fibroblast type (HMFs or CXCL12+HMFs)
(Figure 3.3c-g).
We found that with the softer ECM, the effect of fibroblast cell density was minimal and by day 4, the ECM shrank by 75-88% with HMFs and 88-94% with
CXCL12+HMFs (Figure 3.3e,f). The stiffer ECM showed significantly less contraction than the softer ECM for a given cell density. For example, on day 4 and with 1×104 HMFs, the stiffer ECM showed only 10% contraction, whereas the softer ECM had 87% contraction. With both fibroblasts, there was a cell density- dependent contractility effect, and larger cell densities caused a greater ECM contraction. For example, with the stiffer ECM on day 10, increasing the density of
HMFs from 5×103 to 1.5×104 cells increased ECM contraction from 41% to 66%.
Increasing density of CXCL12+HMFs in this range increased ECM contraction from
52% to 85%. CXCL12+HMFs more significantly contracted the ECM than HMFs at
61 all cell densities used, and this effect was more pronounced with the stiffer ECM.
For example, with a similar 1×104 cell density and on day 10, CXCL12+HMFs caused 78% ECM contraction compared to 56% with HMFs. Figure 3.3g statistically compares ECM contraction by CXCL12+HMFs and HMFs at the largest cell density used. This analysis used for different cell densities showed that
CXCL12+HMFs always generate significantly greater matrix contraction than
HMFs (p<0.05).
3.2.2 Mechanism of ECM Contraction by Fibroblasts
Consistent with our results, fibroblasts in human tumor stroma show the ability to remodel the ECM and contract it [40,209]. To determine whether potential morphological differences between HMFs and CXCL12+HMFs contribute to their ability to generate force and contract the ECM, we conducted whole gel immunofluorescence of fibroblasts stained with β-actin and Hoechst. Confocal imaging showed that unlike HMFs, CXCL12+HMFs had a spindle-like, elongated morphology (Figure 3.4a). CXCL12+HMFs had a five-fold higher cell aspect ratio than HMFs (Figure 3.4b).
To understand the underlying molecular mechanism, we evaluated activity of myosin light chain 2 (MLC2) protein that regulates actomyosin contractility and cell polarity. Quantifying the ratio of active to total protein showed that
CXCL12+HMFs had 1.40-fold higher p-MLC2/MLC2 than HMFs (Figure 3.4c,d).
This was also consistent with the significantly higher motility (Figure 3.5a,b) and
RhoA and ROCK I activity (Figure 3.5c) in CXCL12+HMFs than HMFs. Using a 62
ROCK inhibitor, Y27632, we fully blocked collagen contraction by both fibroblasts
(Figure 3.5d), establishing the RhoA/ROCK/MLC2 signaling as a major mechanism of collagen contraction by fibroblasts. Both fibroblasts dispersed in collagen had a similar level of metabolic activity, indicating that the difference in their contractility was not due to potential differences in their proliferation in culture
(Figure 3.6). Additionally, supplementing CXCL12-conditioned medium to HMFs did not make these cells more contractile (Figure 3.7), negating a role for this chemokine in fibroblast cell contractility. Overall, activities of the signaling molecules that regulate cell-ECM mechnaotransduction support the greater contractility of CXCL12+HMFs than HMFs in the organotypic tumor model.
3.2.3 The Role of Fibroblasts in Matrix Invasion of Cancer Cells
Next, we studied the role of fibroblasts on CXCR4+TNBC cell invasion of
ECM in the organotypic cultures. We used collagen embedded CXCR4+TNBC cell mass as a negative control to evaluate how HMFs and CXCL12+HMFs regulate breast cancer cell invasiveness. In the absence of fibroblasts, CXCR4+TNBC cells invaded the ECM and showed astral-like protrusions from the spheroid (Figure
3.8a). CXCR4+TNBC cells showed contrasting behaviors when cultured with different fibroblasts. While CXCL12+HMFs promoted a significant ECM invasion of
CXCR4+TNBC cells, HMFs suppressed CXCR4+TNBC cell invasiveness (Figure
3.8a). To quantitatively compare cancer cell invasion of ECM among the three models, we normalized the invasion of CXCR4+TNBC cells of each model on a measurement day by its matrix contractility (Figure 3.8b). This compensated the 63 effect of ECM contraction on masking the invasion distance of cancer cells. Cancer cell spheroids in the three models showed a similar morphology after 24 h of incubation. However, significant differences emerged after 3 days. CXCR4+TNBC cells in the model containing CXCL12+HMFs showed a significantly more invasion than the other two models containing HMFs or without any fibroblasts. With
CXCL12+HMFs, invasion of CXCR4+TNBC cells was 2.70-fold higher than that with HMFs, and 1.30-fold higher than that without fibroblasts (Figure 3.8c).
Maintaining the cultures for 5 days increased these differences to 4.0-fold and 1.5- fold, respectively.
To develop a holistic view of 3D distribution of invading CXCR4+TNBC cells in the three models, we quantified CXCR4+TNBC cell numbers in different imaging planes and constructed histograms of position of the cells along the z-axis (Figures
3.9a and 3.10). On day 3, CXCR4+TNBC cells in models including CXCL12+HMFs or without fibroblasts showed similar invasion profiles and a larger number of cells that invaded a greater distance than the model with HMFs (Figure 3.9a). Results from day 5 samples showed a striking difference in the distribution of invading cells between these models, which could not be captured from the commonly used analysis of a collapsed z-stack of images. In the presence of CXCL12+HMFs,
CXCR4+TNBC cells showed the most scattering along the z-direction, whereas
HMFs significantly inhibited invasion of CXCR4+TNBC cells. In the absence of fibroblasts, there was no significant difference in the travel distance of cancer cells in the z-direction from day 3 to day 5. To enable statistical comparison, we
64 averaged the invasion results from each model that confirmed a significant increase in the ECM invasion of CXCR4+TNBC cells from day 3 to day 5 only in the presence of CXCL12+HMFs (Figure 3.9b).
To identify potential molecular mechanisms underlying invasion of
CXCR4+TNBC cells due to signaling with CXCL12+HMFs, we evaluated activities of several prominent protein kinases associated with tumor cell invasiveness.
Results showed that ERK1/2 activity was significantly higher by 1.40-fold in microtissues containing CXCL12+HMFs than those with HMFs (Figure 3.11a and b). In addition, we determined ERK1/2 phosphorylation in monocultures of only fibroblasts or TNBC spheroids in collagen gel. The active levels of ERK1/2 in both fibroblasts were comparable and even lower in TNBC spheroids. These results suggest that CXCL12 signaling with CXCR4+TNBC cells in the organotypic culture accounts for the increased ERK1/2 activity. We did not observe any significant difference in activities of p-Akt or p-PI3K between the two organotypic models
(data not shown).
3.2.4 Inhibition of ECM Invasion of CXCR4+TNBC
We treated the CXCL12+HMFs-containing microtissues with 1 µM of
AMD3100, which is an antagonist of CXCR4. We imaged the microtissues on days
1, 3, and 5 and used the z-projected images to investigate the effect of blocking
CXCL12 – CXCR4 signaling on ECM invasion of CXCR4+TNBC cells. Only after
24 h of incubation, the treatment significantly reduced ECM invasion of cancer cells
(Figure 3.12a,b). While cancer cell invasion of the matrix increased during culture 65 without treatment, blocking CXCR4 signaling maintained the cancer cells as a minimally-invasive spheroid (Figure 3.12a,b). Our molecular analysis showed that blocking CXCR4 signaling significantly reduced p-ERK1/2 levels in the
CXCR4+TNBC:CXCL12+HMFs microtissue (Figure 3.12c). These results further established the role of MAPK pathway in TNBC cell invasiveness and the potential of inhibiting tumor-stromal signaling as a therapeutic approach.
3.3 Discussion
Unlike approaches using co-cultures or Matrigel to form 3D cultures, our high throughput organotypic tumor model both mimics the architecture of solid tumors and reproduces physical and biochemical properties of the stroma. As a proof of concept in this study, we used breast tissue fibroblasts as stromal cells and type I collagen as the primary component of ECM in breast tissue. To demonstrate feasibility of adjusting properties of the system, we developed tumor models that replicated elastic modulus of breast tumors. The 1 mg/ml and 4 mg/ml collagen matrices had elastic moduli of 0.96 and 2.53 kPa to mimic bimodal stiffness of human breast tumors [28,182]. We also adjusted the cellular composition of the model to a 2:1 ratio of fibroblast:cancer cells to mimic advanced breast tumors that have a larger stromal content, which correlates with tumor progression and poorer prognosis [210,211].
We observed significant contraction of the collagen matrix as early as 48 h after forming the models. This ECM contraction was solely due to the dispersed fibroblasts as spheroids of CXCR4+TNBC cells did not shrink the matrix even after 66
2 weeks of culture. Additionally, there was no significant difference in the matrix contraction of fibroblasts when the culture contained or lacked tumor spheroids
(data not shown). This is consistent with the known capability of fibroblasts to remodel the ECM in normal and pathological conditions such as wound healing, fibrosis, and cancer [212,213]. We characterized ECM contraction as a function of density and type of the fibroblasts (HMFs and CXCL12+HMFs), and elastic modulus of collagen gel. Consistent with results from 3D contraction assays
[213,214], our finding suggests that fibroblasts contract softer collagen hydrogels faster than stiffer ones, and that a greater density of fibroblasts expedites the process. More importantly, we found that CXCL12+HMFs contract the collagen matrix more significantly than their normal counterparts, regardless of cell density or elastic modulus of the matrix. We verified that the two fibroblasts in the organotypic model have similar proliferation rates (Figure 3.6), excluding the possibility of potential differences in their proliferative activities causing the difference in their contractility.
Previous studies showed that activated fibroblasts, known as myofibroblasts, enhance contraction of collagen lattices and feature elongated morphology with stress fibers [215–217]. In addition, myofibroblastic stromal cells derived from diseased fibrotic tissue have enhanced collagen contraction than stromal cells derived from normal tissue [218]. We hypothesized that
CXCL12+HMFs may exhibit myofibroblastic morphology responsible for their higher ECM contraction than HMFs. Indeed, CXCL12+HMFs dispersed in collagen
67 hydrogels had an elongated morphology compared to HMFs that showed a branching morphology (Figure 3.4a). Seeding the fibroblasts on the surface of collagen gels also reproduced this morphological difference (Figure 3.13a).
Additionally, the clear difference in the morphology of these two fibroblasts persisted at the same contractile time point. That is, CXCL12+HMFs on day 13 and
HMFs on day 4 had a similar matrix contractility of about 0.75 (Figure 3.3g) but displayed drastically different morphologies (Figure 3.13b). Consistent with this finding, CXCL12+HMFs were more migratory than HMFs in a wound healing assay
(Figure 3.5a,b). Our molecular analysis showed that CXCL12+HMFs dispersed in collagen gels have a significantly greater cell-matrix mechanotransduction activity than HMFs, i.e., higher levels of contractility-associated proteins in the
RhoA/ROCK I/MLC2 pathway (Figure 3.4c,d, Figure 3.5c). This is consistent with studies that show CAFs generate actomyosin-mediated contraction and remodel
ECM through this pathway [219–222]. These results suggest that the different morphologies of HMFs and CXCL12+HMFs are inherent properties of the cells.
To validate the biological relevance of our organotypic tumor model, we investigated the effect of CXCL12-producing fibroblasts on matrix invasion of breast cancer cells by focusing on the CXCL12 – CXCR4 signaling axis. Among different soluble signaling molecules of CAFs, CXCL12 is a prominent chemokine promoting invasion of cancer cells [53,223]. CXCL12 enhanced invasiveness of
CXCR4+TNBC cells significantly and by four-fold compared to HMFs that lacked this chemokine. It is important to note that we normalized invasion of
68
CXCR4+TNBC cells with matrix contractility that was quantified from 2D projected area of collagen gels. However, gels contract volumetrically. Considering volumetric contraction of the matrix would increase invasion of cancer cells in the model containing CXCL12+HMFs even more. For example, approximating the shapes of hydrogels and quantifying invasion of cancer cells in day 5 cultures using the volumetric approach resulted in eight-fold increase in the invasion of cancer cells in the model containing CXCL12+HMFs than that with HMFs. Overall, this is consistent with studies that showed increased cancer cell invasion in microenvironments enriched with CAFs [221,224], and that CAFs promote invasion and metastasis of tumor cells in vitro and in mouse models [225–227].
We found this signaling converges on MAPK pathway and amplifies ERK1/2 phosphorylation in CXCR4+TNBC cells to mediate enhanced invasion of cancer cells, consistent with the established role of ERK in tumor cell migration and invasiveness [228–230]. As a therapy strategy, we demonstrated that blocking
CXCR4 leads to a significant reduction in ECM invasion of CXCR4+TNBC cells.
This is a promising treatment strategy for TNBC disease because the lack of hormone receptors and low HER2 expression renders endocrine and targeted therapies not feasible.
To further delineate the importance of CXCL12 – CXCR4 signaling in cancer cell invasion, we generated organotypic models containing parental TNBC cells and CXCL12+HMFs fibroblasts. Results showed minimal matrix invasion of parental TNBC cells on day 5 of cultures (Figure 3.14). This is because parental
69
TNBC cells do not express detectable levels of CXCR4 receptors [160], highlighting that invasiveness of breast cancer cells depends on both components of this signaling axis. It is important to recognize that in tumor microenvironments, there are temporally dynamic changes in the expression levels of signaling molecules as the disease progresses. That is, the expression of CXCR4 receptor may be induced due to the persistent presence of its chemokine ligand in tumor microenvironments. These changes often occur over several months to years
[231], and capturing them with in vitro systems may not be readily feasible. In addition, in the organotypic CXCR4+TNBC:HMFs model that lacked CXCL12 to initiate CXCR4 signaling, invasion of cancer cells was significantly reduced, and cancer cells remained confined to a spheroid (Figure 3.8). This is consistent with studies that showed normal mammary fibroblasts limit and even reverse malignant transformation of breast epithelial cells, which restore their baso-apical polarity, secretion of a tissue-specific glycoprotein into the acinar lumens, and deposition of basement membrane [232]. Normal fibroblasts were also shown to inhibit proliferation and invasion of malignant cells through processes including changes in ECM architecture and pro-inflammatory molecules [233]. Elucidating inhibitory effects of normal fibroblasts on invasiveness of TNBC cells in our model requires further studies.
Because co-culture spheroids are now routinely used in cancer research, we aimed to compare the ECM invasion of cancer cells in our organotypic model with intermixed co-cultures of fibroblasts and cancer cells embedded in collagen.
70
Results from intermixed co-cultures showed that cancer cells from both
CXCR4+TNBC:HMFs and CXCR4+TNBC:CXCL12+HMFs co-cultures invaded the
ECM without any significant difference (Figure 3.15a,b) [96]. Consistent with incompatibility of cell adhesion molecules of fibroblasts and cancer cells [234,235], our previous study showed that the fibroblasts and CXCR4+TNBC cells segregate from an intermixed spheroid, and that this effect was greater with HMFs than with
CXCL12+HMFs [133]. Considering that intermixed co-cultures do not mimic the architecture of early solid tumors in terms of spatial distribution of cancer and stromal cells, our results suggest that their use for cancer cell invasion studies can lead to misleading information.
Overall, this study underscores the value of our organotypic model that reproduces spatial positioning of cells in solid tumors where cancer cells are bordered by stroma containing dispersed CAFs [45,236]. The modularity of our model allows convenient compositional adjustments, including addition of other stromal cells such as immune and endothelial cells and adjusting physicochemical properties of the ECM by supplementing various ECM elements of tumor microenvironment such as hyaluronan and fibronectin or using functionalized synthetic matrices. The compatibility of our model with microwell plates enables high throughput testing of chemical compounds using off-the-shelf robotic liquid handling tools for cancer drug discovery applications.
3.4 Summary
71
We developed high throughput organotypic tumor microtissues that mimic the structure and composition of solid breast tumors and reproduce dynamic interactions between breast cancer and stromal cells in presence of a natural ECM.
This model allowed us to mechanistically study the role of fibroblasts in contraction of collagen matrix and invasiveness of TNBC cells. We demonstrated that both components of the CXCR4 – CXCL12 tumor-stromal signaling are required to promote ECM invasion of TNBC cells through oncogenic MAPK pathway. We showed that blocking tumor-stromal signaling using a targeted therapeutic molecule may be used as a therapy strategy to inhibit matrix invasion of breast cancer cells. This convenient-to-use and scalable tumor model offers a great potential to study tumor-stromal interactions and as a drug discovery tool in compound screening applications.
72
Figure 3.1: Formation of organotypic culture. (a) Schematic process of formation of a microtissue containing a CXCR4+TNBC spheroid and fibroblasts dispersed in collagen. (b) 3D confocal reconstruction of a microtissue containing CXCR4+TNBC spheroids (blue) and fibroblasts (green). Collagen was not stained. The quantities in the red box of panel (b) represent distances in micrometers.
73
Figure 3.2: Mechanical properties of ECM. (a) Elastic moduli of type I rat tail collagen gels prepared from different protein concentrations maintained in PBS and measured using AFM within 4 h of gel formation. (b) A typical histogram obtained from force curve fitting result for a 4 mg/ml collagen gel. Error bars are standard error of the mean and n=5.
74
Figure 3.3: Characterize ECM contraction by fibroblasts. (a) 3D view of spatial distribution of fibroblasts (green) in collagen. (b) Fibroblasts were dispersed in collagen at a density of 2×103, 4×103, 6×103, or 8×103 cells per well. Each well contained 20 µl of collagen solution. The number of cells in each well was estimated by analyzing each stack of confocal images in MATLAB. Briefly, images from each stack were opened in ImageJ and XY coordinates of the cells were determined in an excel sheet. The coordinates were imported in MATLAB to count the total number of cells. The cells that had similar XY coordinates in four consecutive images were considered redundant and hence counted only once. Each cell XY coordinates usually repeated in two or three consecutive images. (c, d) Phase contrast images show contraction of collagen gels, 1 mg/ml and 4 mg/ml, by fibroblasts at different densities of 5×103 (5k), 10×103 (10k), and 1.5×104 (15k) on day 4 and day 10. Each density indicates the number of cells per well of 384- well plates. (e, f) Time-dependent contractility of collagen gel, 1 mg/ml, and 4 mg/ml, by fibroblasts of different densities (5k,10k, and 15k). (g) Comparison of matrix contractility of 4 mg/ml collagen gels by 15k density of HMFs and CXCL12+HMFs. *p<0.05, n=16. Two-tailed unpaired t-test was used to compare the two experimental groups.
75
Figure 3.4: Mechanism of collagen contraction by fibroblasts. (a) Immunofluorescence staining of β-actin (green) and Hoechst (blue) in HMFs and CXCL12+HMFs dispersed in collagen on day 5 of culture. (b) Box plot of cell aspect ratio ( 푊𝑖푑푡ℎ ) for HMFs and CXCL12+HMFs dispersed in collagen. The boxes 퐻푒𝑖𝑔ℎ푡 represent the 25th and 75th percentiles with the median shown with a horizontal line inside each box. The mean is shown by a cross symbol inside each box. The whiskers represent the 10th and 90th percentiles of the data. Two-tailed Mann- Whitney test was used to calculate a p-value. *p<0.01, n=10. (c) Western blot analysis of expression levels of phosphorylated and total MLC2 in HMFs and CXCL12+HMFs dispersed in collagen. (d) MLC2 activity quantified as p- MLC2/MLC2 from three separate experiments. Data were normally distributed and two-tailed, unpaired t-test was used to calculate a p-value. *p<0.01.
76
Figure 3.5: Characterization of fibroblasts motility (a) Wound healing assay to study migration of HMFs and CXCL12+HMFs. (b) Migration of cells is quantified as gap closure= 1-A2/A1, where A1 is the initial gap area and A2 is the final gap area. *p<0.05 was calculated using two-tailed, unpaired t-test with n=12. (c) Western blot analysis of expression levels of contractility-associated proteins in HMFs and CXCL12+HMFs cultured in collagen with quantified protein expression after normalizing with GAPDH. Data represent three separate experiments. *p<0.05 was calculated using two-tailed, unpaired t-test. (d) Phase contrast images of cultures of fibroblasts dispersed in collagen gels on day 5 showing inhibition of collagen contraction using 10 µM of a ROCK inhibitor (Y27632, Selleckchem).
77
25000 n.s. n.s.
n.s.
) u
. 20000
a
(
y
t
i
s n
e 15000
t
n i
n.s.
e c
n 10000 e
c n.s.
s
e
r o
u 5000
l F
0 1 3 5 7 9 Day
HMFs CXCL12+HMFs Figure 3.6: Metabolic activities of HMFs and CXCL12+HMFs dispersed in collagen gels measured using a PrestoBlue assay. *p<0.05 was calculated using two-tailed, unpaired t-test with n=16.
78
Figure 3.7: Effect of secretome on fibroblast contractility. (a) Collagen gel containing dispersed HMFs supplemented with conditioned medium from monoculture of CXCL12+HMFs. (b) Matrix contractility of HMFs-containing collagen gels with and without CXCL12+HMFs conditioned medium. *p<0.05 was calculated using two-tailed, unpaired t-test with n=16.
79
Figure 3.8: Influence of fibroblasts on ECM invasion of cancer cell. (a) Confocal images of CXCR4+TNBC cell mass in three different microtissues over a 5-day culture. (b) Matrix contractility for CXCR4+TNBC-containing microtissues with and without 1.5×104 fibroblasts. Unpaired t-test was used to calculate the p-values. *p<0.05, **p<0.001, ***p<0.0001, n=10. (c) Normalized invasion of CXCR4+TNBC cells in microtissues with and without fibroblasts. Invasion area values were normally distributed and two-tailed, unpaired t-test was used to calculate the p- values. *p<0.05, **p<0.001, ***p<0.0001, n=8.
80
Figure 3.9: Analysis of ECM invasion of cancer cells along the z-axis. (a) Histograms of invasion of CXCR4+TNBC cells in three different microtissues on day 3 and day 5. Each colored curve in a graph represents pixels area of CXCR4+TNBC cells from one microtissue sample and each data point is obtained from one image of the confocal stack of images of that microtissue. Histograms have non-Gaussian distribution. (b) Normalized 3D spreading in three different tumor models. Normalized invasion area and AUC values were normally distributed. *p<0.05, **p<0.001, and ***p<0.0001 were calculated using two-tailed, unpaired t-test with n=8.
81
Figure 3.10: XZ images showing the lower hemisphere of a CXCR4+TNBC spheroid (green) in microtissues containing dispersed (a) HMFs or (b) CXCL12+HMFs reconstructed from confocal z-stack images.
82
Figure 3.11: Molecular analysis of TNBC invasiveness. (a) Western blot analysis of ERK1/2 phosphorylation in different microtissues. (b) Quantified p-ERK1/2 normalized with t-ERK1/2 in five different microtissues. Data represent three separate experiments. +HMFs and +CXCL12+HMFs indicate organotypic cultures containing a TNBC spheroid and fibroblasts dispersed in collagen. For statistics, t- test was used to compare +HMFs and +CXCL12+HMFs, whereas one-way ANOVA with Tukey’s pairwise comparisons were used to compare CXCR4+TNBC, HMFs, and CXCL12+HMFs. *p<0.05 and ***p<0.0001.
83
Figure 3.12: Inhibition of tumor-stromal interactions. (a) Confocal z-projected images of the CXCR4+TNBC cells in CXCL12+HMFs-containing microtissues without and with AMD3100 treatment. (b) Normalized invasion area for AMD3100- treated and non-treated microtissues. *p<0.01 was calculated using two-tailed, unpaired t-test with n=8. (c) Western blot analysis of effect of AMD3100 treatment on ERK1/2 activity. The graph shows quantified phospho-protein levels after normalizing with t-ERK1/2. Data represent three separate experiments. ***p<0.0001 was calculated using two-tailed, unpaired t-test.
84
Figure 3.13: Morphologies of fibroblasts (a) Phase contrast images of HMFs and CXCL12+HMFs captured 24 h after seeding on collagen gels. (b). Phase contrast images of HMFs and CXCL12+HMFs dispersed in collagen gels on day 13 and day 4, respectively, when they have comparable matrix contractility. HMFs showed a branched morphology, whereas CXCL12+HMFs had an elongated morphology.
85
300 µm
Figure 3.14: Parental TNBC cells in the collagen gel containing dispersed CXCL12+HMFs show minimal invasion on day 5 of culture.
86
Figure 3.15: Co-culture spheroid model. (a) Normalized invasion of CXCR4+TNBC cells from intermixed fibroblast-TNBC co-culture spheroids into collagen matrix. (b) Representative fluorescent images of cancer cells spreading from intermixed co- culture spheroids into the collagen matrix.
87 CHAPTER IV
ORGANOTYPIC BREAST TUMOR MODEL ENABLES MECHANISTIC STUDIES OF TUMOR-STROMAL INTERACTIONS IN TNBC
Breast cancer is the most diagnosed cancer and the second leading cause of cancer mortality among women in the United States [2,237]. Breast tumors that lack expression of estrogen receptor (ER) and progesterone receptor (PR), and amplification of HER2 receptor are known as triple negative breast cancer (TNBC)
[238,239]. Although TNBC accounts for ~15% of all breast cancers, it has a significantly higher rate of relapse, greater metastatic potential, and the lowest 5- year survival rate among all breast cancer subtypes [240,241]. The median survival of patients with metastatic TNBC is only 13.3 months [16]. Current therapeutic approaches for breast cancer include surgery, radiation, chemotherapy, hormonal therapy, and targeted therapy. Unfortunately, endocrine and hormonal therapies are not feasible with TNBC [242,243]. Additionally, targeted therapy options for TNBC patient are very limited and cytotoxic chemotherapy remains the standard treatment. TNBC is a heterogeneous disease composed of 7 different molecular subtypes: basal-like 1, basal-like 2, immunomodulatory, mesenchymal, mesenchymal stem-like, luminal androgen receptor, and unstable [244]. Studies show that these subtypes of TNBC tumors 88 respond poorly and differently to chemotherapy [244,245]. Likewise, chemotherapy did not improve TNBC patients’ pathologic complete response (pCR), which is defined as the absence of invasive cancer in the breast and lymph nodes after treatment [244]. These chemotherapies or limited molecular targeted therapies are largely ineffective due to intrinsic and/or adaptive drug resistance that eventually leads to relapse and metastasis [9]. Therefore, alternative treatment approaches are imperative to improve outcomes for TNBC patients. Recent evidence suggests that besides genetic alterations in tumor cells, the tumor microenvironment (TME) and its interactions with cancer cells (i.e., tumor-stromal interactions) have a major role in conferring therapeutic resistance and disease progression [11,18]. Using the TME as a therapy target is particularly more important for aggressive cancers such as TNBC that have a rich stroma and currently limited targeted therapies [44,188].
The TME comprises of heterogenous populations of cancer cells and different stromal cells such as fibroblasts, immune cells, and endothelial cells embedded in an extracellular matrix (ECM) along with the diffusible growth factors, chemokines, and cytokines [20]. Among non-cancerous cells, cancer-associated fibroblasts (CAFs) are the most abundant stromal cells and a key regulator of tumorigenesis in TNBC [246]. CAFs secrete various paracrine signaling molecules that interact with cancer cells and promote tumor growth, drug resistance, and eventually metastasis [247]. CAFs mediate these biological processes through soluble signaling of various CAFs-derived factors such as epithelial growth factor
(EGF), hepatocyte growth factor (HGF), fibroblast growth factor (FGF), insulin-like
89
growth factor (IGF), and transforming growth factor-β (TGF- β) [43]. For example,
Figure 4.1 shows a schematic of HGF-MET pathway where the ligand, HGF, binds to its receptor, MET, which activates several downstream signaling pathways including MAPK, PI3K/Akt, and STAT3 and promote survival, growth, and invasion of cancer cells. The role of stroma-derived paracrine factors on tumorigenesis is well recognized and several drugs are under clinical trials to target either CAFs, its secreted factors, or their corresponding receptors expressed on cancer cells
[11,58–60].
While much progress has been made in understanding the role of TME in carcinogenesis and development of chemical compounds that target the TME, the success rate of anti-cancer compounds in clinical trials is disappointing [93]. It is widely accepted that lack of predictive and biologically relevant preclinical models is one of the major hurdles to translate bench-side preclinical results into more effective strategies to treat cancer patient [94,248]. Typically, animal models and monolayer (2D) cell cultures are used as preclinical models for both mechanistic studies and testing drugs. Animals models are often immunocompromised, lack human stroma, and inconsistently predict drug responses of patients [249]. In addition, animal models cannot accommodate the high throughput drug testing that is a critical step in preclinical studies to identify potentially effective compounds for further consideration. This is often done with 2D cell cultures that are compatible with culture in microplates, robotics for various steps of drug testing, and available biochemical assays to determine responses of cells to chemical compounds.
90
Nevertheless, cells cultures as a flat layer lack the native 3D architecture of tumors, often display significantly less resistance to drugs than the same cells in 3D cultures, and mostly fail to predict drug responses in vivo [250].
To address this technological gap, various 3D tumor models have been developed to facilitate interactions of cancer cells and the stroma. These include co-culture spheroids of cancer-stromal cells with or without ECM [96,133], microfluidic devices with tumor and stromal compartments [156,157], organoids
[149], and xenografts [78]. While 3D models have been increasingly used to study interactions in the TME and evaluate anti-cancer drugs, majority of these models either fail to accurately model the spatial distribution of different components of the
TME or are incompatible with automated platforms for high throughput drug screening, which is essential for drug discovery [112,251,252]. We previously developed a 3D organotypic breast tumor model that contains key components of breast TME, i.e., a mass of cancer cells, activated fibroblasts, and ECM, and is compatible with automated liquid handling operations [253].
To demonstrate the utility of this tumor model for mechanistic studies of the
TME, we conducted a comprehensive study to examine effects of interactions between patient-derived CAFs and TNBC cells on tumorigenic functions of cancer cells. We found CAFs and TNBC cells predominantly communicate via HGF-MET paracrine signaling axis, i.e., HGF secreted by CAFs activates MET on TNBC cells.
This signaling led to activities of several oncogenic pathways in TNBC cells not previously active, including MAPK, PI3K/Akt, and STAT3. We found that CAFs
91 significantly promoted proliferation, invasiveness, and epithelial-to-mesenchymal transition (EMT) in TNBC cells. Our preliminary studies to target this tumor-stromal interaction shows the feasibility of identifying drug combinations that significantly block these tumorigenic activities of TNBC cells. In addition, we found that the
HGF-MET signaling axis was also highly active in the lung stromal environments, which is a common site of TNBC metastasis, and promoted the colonization of
TNBC cells. Disrupting tumor-stromal interactions also effectively reduced colony formation capacity of TNBC cells in lung stroma, indicating the feasibility of identifying effective treatments against both primary and metastatic tumors.
Altogether, our studies demonstrated the potential of the organotypic tumor model to mechanistically understand tumor-stromal interactions and develop new therapeutic strategies.
4.1 Materials and Methods
The methods for phospho-RTK array, bead-based multiplex immunoassay, confocal microscopy of organotypic tumor models, quantitative PCR, combination treatments of organotypic cultures, and colony formation assay are described below.
4.1.1 Cell Culture
MDA-MB-231, SUM159, Hs 578T, MDA-MB-157, HCC1806, and BT-20
TNBC cells, and WI-38 lung fibroblast cells were purchased from ATCC. MDA-MB-
231 and SUM159 TNBC cells transduced to express an endogenous green fluorescent protein (GFP) were a generous gift of Dr. Gary D. Luker (University of 92
Michigan). Three different fibroblast cells were used: normal human mammary fibroblasts stably transduced with mCherry protein (labeled HMF) (gift of Dr. Daniel
Hayes, University of Michigan), cancer-associated fibroblasts derived from a human breast carcinoma tumor (labeled CAF-1) and obtained from Neuromics
(Cat. No. CAF06), and cancer-associated fibroblasts derived from an invasive ductal breast carcinoma tumor (labeled CAF-2) and obtained from BioIVT (Cat.
No. HPCCAFBR-05). CAFs were cultured as recommended by the vendors for up to seven passages only. MDA-MB-231, MDA-MB-157, Hs 578T, and HMF cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Sigma) supplemented with 10% fetal bovine serum (FBS, Sigma), 1% glutamine (Life
Technologies), and 1% antibiotic (Life Technologies). HCC1806 was cultured in
RPMI 1640 medium (Sigma) supplemented with 10% FBS and 1% antibiotic.
SUM159 cells were cultured in Ham’s F-12 medium (Gibco) supplemented with
10% FBS, 5 µg/ml insulin (Sigma), 2 µg/ml hydrocortisone (Sigma), 1% glutamine
(Life Technologies), and 1% antibiotic. WI-38 cells were cultured in Eagle’s
Minimum Essential Medium (EMEM, ATCC) supplemented with 10% FBS and 1% antibiotic. Cells were plated in T75 flasks (Thermo Fisher Scientific) and kept in a humidified incubator at 37C and 5% CO2. When the TNBC cells were about 80% confluent, they were rinsed with phosphate buffered saline (PBS, Sigma), dislodged using 0.25% trypsin (Life Technologies), and sub-cultured.
4.1.2 Human phospho-RTK array
93
A human phospho-RTK array (R&D Systems, Cat. No. ARY001B) was used to simultaneously detect phosphorylation of 49 different human receptor tyrosine kinases (RTKs) on TNBC cells after a brief stimulation with conditioned medium from cultures of CAF cells. Briefly, TNBC cell lines (MDA-MB-231 and SUM159) were separately stimulated for 10 min with conditioned medium from a confluent culture of CAF cells (CAF-1 and CAF-2). For negative control, TNBC cells were incubated with culture medium only for 10 min at 37°C. The cells were rinsed with cold PBS and lysed using lysis buffer 17 containing protease inhibitors (Aprotinin,
Leupeptin, and Pepstatin). Protein concentrations of the samples were quantified using a BCA protein assay kit (Thermo Fisher Scientific). Each array was blocked for 1 h with array buffer 1 and incubated with 300 µg of protein lysate overnight at
4°C. The array was washed and incubated with HRP conjugated anti-phospho- tyrosine detection antibody and developed using FluorChem E imaging system
(ProteinSimple). The relative phosphorylation of each RTK in each blot was quantified by measuring the pixel densities of the corresponding dot on the array using an image analysis software (ImageJ) and normalized with respect to the corresponding vehicle control group.
4.1.3 Bead-Based Multiplex Immunoassay
Conditioned medium from monocultures of HMF, CAF-1, CAF-2, and WI-38 cells were collected. Growth factors secreted by these cells were quantified using commercially available kits. Growth factors HGF, FGF-1, FGF-2 were analyzed using a growth factor magnetic bead panel (Millipore Sigma, Cat. No. 94
HAGP1MAG-12K) and EGF and PDGF-AB/BB were analyzed using another growth factor magnetic bead panel (Millipore Sigma, Cat. No. HCYTA-60K), according to manufacturer’s instructions. Briefly, the conditioned media were centrifuged to remove debris. Then, the supernatant was incubated with the premixed antibody-immobilized beads overnight at 4°C on a titer plate shaker. The beads were washed carefully three times with a wash buffer using a handheld magnet (Millipore, Cat. No. 40-285) to retain the magnetic beads in the well plate.
The beads were incubated with a detection antibody followed by addition of streptavidin-phycoerythrin. Median fluorescent intensity (MFI) from the beads was measured using a Luminex MAGPIX instrument [254,255]. Standard curves for the soluble factors were generated to determine concentrations of the soluble factors.
4.1.4 Organotypic Tumor Model Formation
The organotypic tumor model was formed in two steps according to our established protocol [253]. First, spheroids of TNBC cells were formed at a density of 7.5×103 cells per microwell in an ultralow attachment 384-well plate (Corning) using an aqueous two-phase system [128,129]. After TNBC spheroids formed, medium was robotically removed from the wells and 20 µl of ice-cold rat tail type I collagen solution (Corning) containing 1.5×104 suspended fibroblast cells was dispensed into each well. An in-house cooling system was used to maintain collagen solution at 4°C and prevent premature gelation of collagen. The cooling system comprised of Peltier plates arranged in series at the bottom of well plates, a digital temperature relay to control the temperature of the Peltier plates powered 95 by a 12V DC supply, and an aluminum water cooling block under the hot side of the Peltier plate to dissipate the heat produced during the cooling process. The
384-well plate was incubated at 37°C for 30 min to form a collagen gel and then
60 μl of medium containing only 1% FBS was added to each well.
4.1.5 Confocal Microscopy of TNBC Cell Invasion in Organotypic Tumor Models
For the imaging purpose, the 3D cultures were formed in a glass bottom
384-well plate (MatTek, Cat. No. PBK384G). Images were captured using a confocal microscope (Nikon A1) and with a 10X objective on days 1 and 5 of culture to assess matrix invasion of cancer cells. A 488 nm laser was used to excite the fluorophore and a 500–550 nm filter was used to capture the GFP fluorescence of
TNBC cells. The TNBC cells without endogenous GFP were stained with a Calcein
AM dye (Thermo Fisher Scientific) for confocal imaging. Z-stacks were constructed from images of samples acquired with a z-spacing of 20 μm. NIS software was used for image acquisition and Fiji (ImageJ, NIH) was used for the analysis and
3D reconstruction of the images. TNBC cell invasion of the ECM was quantified by analyzing images from z-projected images [253]. The total pixel area of TNBC cells when cultured with CAF-1 and CAF-2 was normalized to the pixel area of TNBC cells when cultured with HMF. To visualize unlabeled collagen fibers using confocal reflectance microscopy, the dichroic mirror was set to BS20/80 to allow light transmission, the fourth channel was set up for reflection using the 561 nm laser, and all channel light paths were set to ‘through’ in the software options [256].
96
4.1.6 Flow Cytometry of Organotypic Cultures
To quantify the proliferation of TNBC cells in the organotypic model, the cultures were maintained for 5 days at which point the collagen matrix was digested by treating the cultures with collagenase I (Sigma) for 15 min along with mechanical agitation using pipetting. Collagenase activity was neutralized by adding triple amounts of complete growth medium using the manufacturer's protocol. The resulting suspension was centrifuged to obtain a cell pellet that contained both TNBC and fibroblast cells. The cell pellet was suspended in 300 µl of medium and then 50 µl of suspension of counting beads (CountBright,
Invitrogen) was added. Cells were analyzed with a BD Accuri C6 flow cytometer.
At least 1000 counting bead events were acquired while running each sample.
Unstained TNBC cells that lacked GFP were used to remove the background fluorescence. The GFP+ TNBC cells (MDA-MB-231 and SUM159) were gated from unstained fibroblasts and the absolute TNBC cell number was calculated using:
TNBC count = (Number of TNBC events/number of bead events) × number of beads
4.1.7 Quantitative PCR
Cells were harvested from organotypic cultures and lysed using a Total
RNA Kit (TRK) lysis buffer (Omega BioTek). The lysate was homogenized by passing it through homogenizer mini columns (Omega BioTek). Total RNA was obtained using an RNA isolation kit (Omega BioTek). After removing DNA using
RNase-free DNase (Omega BioTek), purity and concentration of isolated RNA 97 were assessed using optical density (OD) 260/280 spectrophotometry (Synergy
H1M, BioTek Instruments). cDNA was synthesized from 1 μg of total RNA using random hexamer primers (Roche). Real-time qPCR was performed in a
LightCycler 480 instrument II using a SYBR Green Master Mix (Roche). After combining 50 ng of cDNA with the primer and the SYBR Green Master Mix to a final volume of 15 μl, the reactions were incubated at 95°C for 5 min followed by
45 cycles of amplification, that is, at 95°C for 10 s, at 60°C for 10 s, and at 72°C for 10 s. The primer sequences for the genes are listed in Table 4.1. Expression levels of different proliferation gene markers were calculated relative to β-actin and
GAPDH using the ΔΔCt method. The fold change in mRNA expression of each gene was determined according to the 2−ΔΔCt method [257,258].
4.1.8 Western Blotting
For organotypic cultures, cells were harvested using collagenase I treatment for 15 min until the matrix was solubilized. Collagenase activity was neutralized by adding triple amounts of complete growth media. The resulting suspension was centrifuged to obtain a cell pellet. The cells were washed with cold
PBS and lysed with 500 µl of complete RIPA buffer containing 1% protease inhibitors and 1% phosphatase inhibitors.
For stimulation experiments, TNBC cells (MDA-MB-231 and SUM159) were separately stimulated for 10 min with conditioned medium from confluent CAF cells cultures (CAF-1 and CAF-2) with or without MET inhibitor, crizotinib (0.5 µM or 1
98
µM). For negative control, TNBC cells were incubated with culture medium only for 10 min at 37°C. The cells were washed with cold PBS and harvested using
RIPA buffer.
The cell suspension was further sonicated twice at a 20% amplitude for 5 sec (Vibra-Cell, Sonics). A BCA protein quantification assay was used to measure total protein concentration from each sample, which was then loaded onto a 4-15% gel (Bio-rad) for electrophoresis. The gel was transferred onto a nitrocellulose membrane by electroblotting, blocked in 5% non-fat dried milk prepared in wash buffer for 1 h, and incubated overnight with primary antibodies. Primary antibodies used were phospho-p44/42 MAPK (Erk1/2, Cat. No. 9101), p44/42 MAPK (Erk1/2,
Cat. No. 9102), phospho-Akt (Ser473, Cat. No. 4060), Akt (pan, Cat. No. 4691), phospho-Stat3 (Tyr705, Cat. No. 9145), Stat3 (Cat. No. 4904), E-Cadherin (Cat.
No.14472), N-Cadherin (Cat. No. 13116), Vimentin (Cat. No. 5741), phospho-Met
(Tr1234/1235, Cat. No. 3077), Met (Cat. No. 8198), and GAPDH (Cat. No. 5174), all purchased from Cell Signaling Technology. The membranes were washed and then incubated with horseradish peroxidase (HRP)-conjugated secondary antibody for 1 h. Detection was done using ECL chemiluminescence detection kit
(GE Healthcare) with FluorChem E imaging system.
4.1.9 Drug Treatments
MAPK pathway inhibitors (trametinib, ulixertinib, and SCH772984) and
PI3K pathway inhibitors (dactolisib, apitolisib, and VS-5584) were purchased from
99
Selleckchem. Dactolisib was dissolved in dimethylformamide and all other compounds were dissolved in dimethyl sulfoxide (DMSO). All compounds were tested dose dependently against free floating spheroids of MDA-MB-231 and
SUM159 at concentrations of 1 × 10−4 μM, 1 × 10−3 μM, 5 × 10−3 μM, 1 × 10−2 μM,
1 × 10-1 μM, 1 × 100 μM, and 1× 101 μM. Untreated spheroids were grown in drug- free cell culture medium. After 4 days of drug treatment, the GFP signal from TNBC cells was measured using a plate reader at excitation and emission wavelengths of 485 and 530 nm, respectively. The fluorescent signal from drug-treated spheroids was normalized with that from the control spheroids and used to construct dose−response curves (GraphPad Prism). A 50% lethal dose (LD50) value was obtained from the dose−response curve of each compound.
4.1.10 Cyclic Treatment of TNBC Spheroids
Adaptive resistance of MDA-MB-231 cells to single-agent therapies was modeled using trametinib (MEK inhibitor) [115]. Co-culture spheroids of MDA-MB-
231 and CAF-1 cells (1:1) were cyclically treated with 5 nM trametinib (i.e.,
0.5×LD50 concentration of trametinib). Each experiment included four cycles of treatment each followed by a recovery phase, and each phase lasted 4 days. The treatment and recovery phases were denoted as T1, T2, T3, T4 and R1, R2, and
R3, respectively. The drug-containing medium was thoroughly removed before adding fresh medium containing conditioned medium of CAF-1 cells. Phase contrast images were captured at the end of each phase to approximate the volume of spheroids. The GFP signal from TNBC cells was measured with a plate 100 reader and normalized with that from day 0 and represented as a percent viability.
A growth rate metric (kc), defined as the difference in the volume of spheroids after and before each treatment divided by the number of days of treatment (4 days), was used to quantify resistance of spheroids to trametinib.
4.1.11 Combination Treatments of Organotypic Cultures
Organotypic cultures were formed using TNBC cells (MDA-MB-231 and
SUM159) and CAFs and treated with crizotinib/trametinib or crizotinib/dactolisib combinations prepared in culture medium with only 1% FBS. Each inhibitor was used at five different concentrations, ranging from 0.1 nM to 50 nM for trametinib and dactolisib, and 1 nM to 5 µM for crizotinib. These concentrations were selected based on our preliminary dose-dependent drug treatment study of free-floating spheroids of these cells. Single-agent treatments with trametinib and dactolisib were also carried out in parallel, resulting in a 6×6 matrix of concentration pairs.
Each concentration used four replicates. Confocal images of the cultures were captured on day 1 and day 5 of incubation. Treatments were renewed after 72 h of incubation. The area occupied by the TNBC cells was measured, normalized with the respective vehicle control, and represented as an invasion area to quantify the effect of each treatment. The fraction of cells affected by each treatment was calculated as (1-invasion area). A synergy analysis was performed to generate a combination index (CI) for several pairs of drugs [259].
101
4.1.12 Colony Formation Assay and Immunofluorescence Imaging of Proliferative TNBC Cells
TNBC cells (MDA-MB-231, MDA-MB-157, SUM159, and Hs 578T) were seeded as a single cell suspension in a 384-well plate with or without lung fibroblast cells (WI-38) in a 1.2% (w/v) methylcellulose gel made using medium containing
1% FBS only. A 1:4 ratio of TNBC:WI-38 cells was used. Each well contained 50
TNBC cells and 200 WI-38 cells. Culture medium with 1% FBS was added after 2 days of seeding in the methylcellulose gel and the medium was replaced every 3 days. The cultures were treated with a MET inhibitor, crizotinib, at 0.5 µM and 1
µM concentrations, and maintained for 8-10 days to image the colonies using a fluorescent microscope. The TNBC cells without GFP were incubated with a cell- permanent dye, Calcein AM, prior to imaging at a 2.5X magnification. An in-house python code in ImageJ was used to automatically detect the colonies in each image and compute the areas of the colonies. A threshold colony diameter of 75
µm was used for statistical analysis.
The TNBC colonies were fixed using 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature and washed with PBS for 10 min. The cultures were blocked with 3% donkey serum for 1 h at room temperature. Cells were permeabilized using 0.3% Triton X-100 in PBS. Samples were incubated with a Ki-
67 rabbit antibody prepared in 1% BSA and 0.3% Triton X-100 in PBS overnight
(Cell Signaling, Cat. No. 9027). Samples were washed and then incubated with a rhodamine red donkey anti-rabbit IgG (Jackson ImmunoResearch, 1:100, Cat. No.
711-295-152) for 1h and imaged using a confocal microscope. Ki-67+ cells were
102 counted per TNBC colony and Ki-67/colony area ratios were further normalized to the average in the respective control group, i.e., TNBC cells without WI-38 cells and crizotinib treatment.
4.1.13 Bioinformatics Analysis of Data from Breast Cancer Patients
The mRNA expression data from 2509 publicly available cases of invasive breast carcinoma were downloaded from The Cancer Genome Atlas (TCGA,
METABRIC dataset) [260], using the cBioPortal portal (http://www.cbioportal.org/)
[261]. A query was performed using EGFR and MET genes and their co- expression was analyzed. The co-expression protein profile of EGFR and MET was also analyzed using protein expression data from the Firehose legacy breast cancer dataset (TCGA). The activity profile of phosphorylated EGFR (p-EGFR) and MET (p-MET) was obtained from The Cancer Proteome Atlas (TCPA) available through MD Anderson Cancer Center [262].
4.1.14 Gene Set Enrichment Analysis (GSEA)
Gene expression profiles of GSE138139 were downloaded from the GEO database [263]. GSEA was performed with a desktop version of the GSEA software (v.4.1.0) [264]. The number of permutations was set to 1,000. The platform of this dataset is the GPL1261 Affymetrix mouse genome 430 2.0 array.
The enrichment score was calculated according to metastasis status (lung metastasis vs. parental breast cancer cells). Differences of a nominal p<0.05 and an FDR less than 25% were defined as significant. 103
4.1.15 Statistical Analysis
Data were first checked for normality using the Anderson-Darling method in
MINITAB. For normally-distributed data, one-way ANOVA with Tukey’s pairwise comparisons were used to compare means among three or more samples. Two- tailed, unpaired t-test was used to compare two experimental groups. Normally- distributed data from experiments were expressed as mean ± standard error. Non-
Gaussian distributed data were analyzed using two-tailed Mann-Whitney test.
Values of p<0.05 denoted statistical significance.
4.2 Results
Below, the results of CAFs-TNBC signaling analysis, activation of pro- metastatic functions in TNBC cells, design of combination treatments to suppress pro-metastatic functions in TNBC cells, and the role of HGF-MET pathway in lung metastatic environment are presented.
4.2.1 CAFs Interact with TNBC Cells via HGF-MET Pathway
To identify which receptors on TNBC become activated due to signaling with CAFs derived from breast cancer patients, we stimulated TNBC cells with conditioned medium from CAF-1 and CAF-2 cultures and used a phospho-RTK array to detect activities of 49 RTKs in TNBC cells. Figure 4.2a shows results with
MDA-MB-231 and SUM159 TNBC cells. RTKs other than EGFR and MET did not have significant fold change after CAFs stimulation (Figure A-1). While both TNBC cells had high basal levels of EGFR activity, stimulation with CAFs resulted in 104 significant activation of MET. CAF-1 led to 25 and 30-fold increase in pMET levels and marginally affected pEGFR by 0.7 and 1.5-fold in MDA-MB-231 and SUM159 cells, respectively (Figure 4.2b). CAF-2 also increased pMET by 3.7 and 5.5-fold and minimally changed pEGFR by 0.8 and 1.3-fold in MDA-MB-231 and SUM159 cells, respectively. These results indicated that among the 49 RTKs, CAFs predominantly activate MET in TNBC cells. To further validate this finding, we performed Western blotting of six different TNBC cells treated with conditioned media from the CAFs. Results showed significant activation of MET in all six TNBC cell lines by CAF-1 and in five of the TNBC cell lines by CAF-2 (Figure 4.2c,d).
CAF-1 generated a more pronounced effect on MET activity than CAF-2 did.
Despite MET protein expression in the TNBC cells (i.e., high t-MET levels), only paracrine signaling with CAFs activated it (i.e., p-MET) (Figure 4.2c). To further validate this finding, we performed an immunoassay to determine prominent signaling molecules secreted by the CAFs. Both CAFs secreted the MET ligand, hepatocyte growth factor (HGF), significantly more than normal HMF cells did, i.e.,
~10.2 ng/ml by CAF-1, ~1 ng/ml by CAF-2, and only 50 pg/ml by HMF (Figure
4.2e). Other soluble factors such as epidermal growth factor (EGF), fibroblast growth factors (FGF-1 and FGF-2), and platelet-derived growth factors (PDGF
AA/AB) were not detectable. Taken together, these results strongly indicate that
CAFs from patients secrete HGF and predominantly activate MET in TNBC cells.
4.2.2 CAFs Activate Multiple Downstream Pathways in TNBC Cell Lines
105
Active RTKs regulate various signaling pathways that drive breast cancer progression [265,266]. To identify pathway level effects of CAFs-mediated HGF-
MET signaling in TNBC cells, we evaluated activities of several prominent pathways in breast cancers including MAPK, PI3K/Akt, and JAK/STAT. While
MDA-MB-231 cells had basal MAPK/ERK activity, CAF-1 led to activation of both
Akt and STAT3, whereas CAF-2 only activated Akt (Figure 4.3a). In SUM159 cells that had basal PI3K/Akt pathway activity, both CAFs activated ERK and STAT3 signaling (Figure 4.3a). CAF-1 had a greater effect in activation of these pathways than CAF-2 because of its ~10-fold higher HGF production (Figure 4.2e) that led to significantly higher MET activation in TNBC cells (Figure 4.2c,d). To validate that the activation of these pathways was indeed mediated by CAFs through the
HGF-MET axis, we used a MET inhibitor, crizotinib, against TNBC cells treated with conditioned medium of CAF-1. We selected only CAF-1 for this experiment because of its greater effect on promoting oncogenic signaling in TNBC cells. Both
MDA-MB-231 and SUM159 cells showed a dose-response to crizotinib (10 min treatment) and displayed a significant reduction in phosphorylated levels of the signaling molecules ERK, Akt, and STAT3 (Figure 4.3b). Overall, these results establish that CAFs-mediated activation of MET and several downstream pathways in TNBC cells is due to HGF-MET signaling.
4.2.3 MET Expression and Activity in Patient Tumors
To determine the clinical relevance of MET in TNBC, we first analyzed MET and EGFR gene expression in different breast cancer subtypes in TCGA. Co- 106 expression of MET and EGFR was predominant in basal and claudin-low subtypes that primarily represent TNBC disease (Figure 4.4a). Additionally, our analysis of mRNA expression of MET and EGFR in breast cancer showed significantly greater expression levels in basal and claudin-low cohorts than in other breast cancers
(Figure 4.4b). Co-expression of MET and EGFR in clinical samples was strongly correlated in both gene and protein levels (Figure 4.4c). To determine functional activities of MET and EGFR in breast tumors, we analyzed data for invasive breast carcinomas and found a strong correlation between p-MET and p-EGFR (Figure
4.4d). Collectively, this set of analysis of data from patient tumors underlines the significance of MET activity in TNBC, and together with our results, suggests a critical role for CAFs to predominantly activate MET with EGFR+ TNBC cells.
4.2.4 CAFs Promote Pro-Metastatic Functions in TNBC Cells
Our organotypic tumor model mimics the architecture and positioning of cancer and stromal cells in solid tumors and has a matrix stiffness of ~2.5 kPa as in breast tumors (Figure 4.5a) [182,253]. Using this model, we studied the role of
CAFs on matrix invasion of TNBC cells. We found that unlike HMF cells, CAF-1 and CAF-2 promoted ECM invasion of MDA-MB-231 cells by 1.9 and 2.6-fold and
SUM159 cells by 1.6 and 2.3-fold, respectively (Figure 4.5b,c). Our cell count analysis of organotypic models showed that both CAFs also significantly promoted proliferation of TNBC cells. CAF-1 and CAF-2 led to 1.3 and 1.2-fold increase in
MDA-MB-231 cells and 1.2 and 1.1-fold increase in SUM159 cells, respectively
(Figure 4.5d,e). 107
Due to the role of epithelial-to-mesenchymal transition (EMT) in cancer cell invasion and metastasis [267], we investigated whether interactions with CAFs regulate an EMT phenotype in TNBC cells. In organotypic models, we observed morphological differences in invading TNBC cells when the cultures contained
CAF-1 or CAF-2 compared to those containing HMF cells (Figure 4.6a). We quantified cell aspect ratio of invading cells in these three conditions and found that both CAF-1 and CAF-2 promoted a significantly greater mesenchymal morphology in both TNBC cells than HMF did (Figure 4.6a). To validate the results from the morphological analysis, we harvested cells from organotypic cultures and quantified the expression of prominent gene markers of EMT transcription factors.
A majority of the EMT markers in TNBC cells were upregulated due to interactions with CAF-1 and CAF-2 (Figure 4.6b). CAF-1 promoted SNAIL and FOXC2 in MDA-
MB-231 cells and SNAIL and ZEB2 in SUM159 cells by more than 3-fold. CAF-2 also promoted SLUG and FOXC2 by nearly or more than 2-fold in MDA-MB-231 cells and SNAIL in SUM159 cells by over 2-fold. We further validated these results using a Western blot analysis that showed CAF-1 led to elevation of vimentin in
MDA-MB-231 cells and N-cadherin in SUM159 cells (Figure 4.6c). CAF-2 cells did not induce a detectable change in the expression of these proteins in the TNBC cells. This is possibly because SUM159 cells had a high basal level of these proteins and that CAF-2 had less of an effect on TNBC cells than CAF-1 did.
Overall, these gene and proteins expression results suggest that CAFs promote
EMT in TNBC cells.
108
4.2.5 Modeling Resistance of TNBC Cells to Single-Agent Treatments
Having established that CAFs promote oncogenic signaling in TNBC cells, we investigated the effectiveness of therapeutic targeting of the respective pathways in the TNBC cells. To select drugs for these experiments, we used MDA-
MB-231 and SUM159 spheroids and performed a dose-dependent screening of a set of molecular inhibitors including trametinib, ulixertinib, and SCH772984 against the MAPK pathway and dactolisib, apitolisib, and VS-5584 against the PI3K/Akt pathway (Figure 4.7). Our quantitative analysis showed that trametinib was the most effective inhibitor against MDA-MB-231 cells with an LD50 of 10 nM, whereas dactolisib was the most effective inhibitor against SUM159 cells with an LD50 of 1
µM. As a proof-of-concept study to show that TNBC cells become resistant to single-agent treatments and justify the need for combination treatments, we formed co-culture spheroids of MDA-MB-231:CAF-1 and treated them with trametinib, following a cyclic treatment/recovery regimen (Figure 4.8a). We selected this treatment regimen to mimic how patients receive chemotherapy
[259]. Trametinib shrank the spheroids during the first two treatment rounds (T1 and T2) (Figure 4.8b) and reduced the cancer cell viability to 41% at the end of T2
(Figure 4.8c). However, after the second recovery round (R2), trametinib became significantly less effective. At the end of the 28-day regimen, spheroids treated with trametinib were 3.04-fold larger and the GFP fluorescence intensity of the MDA-
MB-231 cells, i.e., an indirect and approximate measure of number of cells, increased by 3.11-fold than those at the end of treatment T1. We used a growth
109 rate metric (kc) to quantify the effects of treatments (Figure 4.8d). Although kc for the spheroids reduced during T1 and T2, it significantly increased during the subsequent treatments T3 and T4. The kc values of spheroids from T1 to T4 significantly increased from -0.0059 mm3/day to 0.0172 mm3/day for the trametinib treatment. The significant decrease in the efficacy of trametinib during the cyclic regimen indicates that TNBC cells develop resistance to treatment with the MEK inhibitor. This is consistent with the frequent failure of single-agent treatments of solid tumors that may occur due to activation of compensatory feedback mechanisms or activation of multiple tyrosine and serine/threonine kinases
[268,269], and necessitates alternative treatment strategies such as combination drug treatments that have been pursued for various cancers.
4.2.6 Treatment Strategies to Suppress Pro-Metastatic Functions of TNBC Cells
Next, we evaluated the feasibility of a combination treatment strategy to suppress various pro-metastatic processes of TNBC cells in organotypic cultures.
We hypothesized that since HGF secretion by CAF cells activated MET in TNBC cells, treatments that block the HGF-MET axis and its downstream signaling pathways should inhibit the resulting tumorigenic activities of TNBC cells. To test this hypothesis, we selected a combination of inhibitors of MET and MAPK pathway against MDA-MB-231 cells and a combination of inhibitors of MET and
PI3/Akt pathway for SUM159 cells. This selection reflects different constitutively active kinase pathways in the TNBC cells, i.e., MAPK pathway in MDA-MB-231
110 and PI3K/Akt pathway in SUM159 cells (Figure 4.3a). Again, we selected only
CAF-1 for subsequent experiments because of its greater effect on promoting oncogenic signaling in TNBC cells.
We first performed matrix-style combination treatments of the organotypic cultures to study effects on inhibition of invasiveness of TNBC cells. Figure 4.9a shows the representative images of the MDA-MB-231 spheroids in an organotypic cultures after single-agent and combination treatments using crizotinib and trametinib. The vehicle control (no treatment) had the maximum invasion area. As a single-agent treatment, both crizotinib and trametinib dose-dependently reduced the invasiveness of MDA-MB-231 spheroids with trametinib being more effective than crizotinib. For example, treating the organotypic cultures with 5 µM crizotinib reduced the invasion area of MDA-MB-231 spheroids by 68% compared to the vehicle control, while treatment with 50 nM trametinib reduced the invasion area by 97%. Majority of the combination treatments blocked invasiveness of MDA-MB-
231 spheroids more than the corresponding single-agent treatments. Among the
25 combinations, we selected two combinations that reduced TNBC invasiveness significantly more than the corresponding single-agent treatments with a high synergy, i.e., CI<0.5. The 1 µM/1 nM and 1 µM/5 nM crizotinib/trametinib pairs had
CI values of 0.26 and 0.34 and were highly synergistic against TNBC invasiveness.
For follow-up molecular analysis, we selected the 1 µM/5 nM crizotinib/trametinib pair because using higher concentration of trametinib also ensures significant
111 killing of MDA-MB-231 cells as evident from the dose-response of MDA-MB-231 spheroids to trametinib (Figure 4.7a).
To study inhibitory effects of 1 µM/5 nM crizotinib/trametinib pair on downstream signaling pathways, we performed Western blot analysis of MDA-MB-
231 organotypic cultures. In parallel, we performed single-agent treatments at the same concentration of each inhibitor used in the combination treatment. Results showed that crizotinib/trametinib pair significantly downregulated all three MAPK,
PI3K/Akt, and JAK/STAT pathways, as shown respectively by ERK, Akt, and
STAT3 bands (Figure 4.9b,c). This effect was significantly greater than those from the respective responses when each drug was used alone. Next, we studied the effect of crizotinib/trametinib pair on proliferation of MDA-MB-231 cells. Using flow cytometry analysis of organotypic cultures, we found that this pair significantly reduced the proliferation of MDA-MB-231 cells than in the vehicle controls or single-agent treatments (Figure 4.9d). We also investigated effects of combination treatment on gene expression of EMT markers in TNBC cells and found that the combination treatment significantly reduced EMT markers such as SLUG, FOXC2, and TWIST, compared to the vehicle control and single-agent trametinib treatments (Figure 4.9e).
With SUM159 organotypic cultures, Figure 4.10a shows representative images after single-agent and combination treatments using crizotinib and dactolisib. The vehicle control (no treatment) had the highest invasion area.
Treating the organotypic cultures with 5 µM crizotinib reduced the invasion area of
112
SUM159 spheroids by 81% compared to vehicle control, while using 5 µM dactolisib reduced the invasion area by 96%. Among the 25 combination treatments, we selected two combinations that reduced TNBC invasiveness significantly more than the corresponding single-agent treatments with a high synergy, i.e., CI<0.5. The 1 µM/100 nM and 1 µM/1 µM crizotinib/dactolisib pairs had CI values of 0.30 and 0.39 and were highly synergistic against SUM159 cell invasion. For follow-up molecular analysis, we selected the 1 µM/1 µM crizotinib/dactolisib pair to also ensure a significant killing of SUM159 cells as indicated by the dose-response of SUM159 spheroids to dactolisib (Figure 4.7b).
Results from Western blotting showed that while 1 µM dactolisib alone was effective in reducing activities of all three ERK, Akt, and STAT3 signaling in
SUM159 cells, the combination treatment was significantly more effective against activities of these signaling pathways (Figure 4.10b,c). Dactolisib alone and crizotinib/dactolisib pair significantly reduced proliferation of SUM159 cells compared to the vehicle controls or crizotinib treatment only (Figure 4.10d). The 1
µM/1 µM crizotinib/dactolisib pair significantly reduced SNAIL, E47, and TWIST
EMT markers in SUM159 cells, compared to dactolisib only (Figure 4.10e). Overall, these results suggest that CAF-mediated pro-metastatic functions of TNBC cells were suppressed using a combination drug pair tailored toward specific TNBC cell types.
4.2.7 The Role of HGF-MET Pathway in Lung Metastatic Environments
113
Lung is a common site of TNBC metastasis. TNBC cells have a higher likelihood to metastasize to lung than bone or liver [24,25]. Studies show that the median time to recurrence of metastatic tumors is shortest in lung and TNBC with lung metastasis has the worst survival than liver or bone metastases [270]. Thus, we asked whether HGF-MET pathway has a role in supporting TNBC colonization in lung. We used a Molecular Signatures Database (MSigDB) hallmark gene set collection and derived 14 MET signature genes based on their association with activated MET. Our analysis of results from a study of murine metastasis model of human TNBC showed that these genes were significantly enriched in lung metastasis (Figure 4.11a) [271]. This was consistent with separate studies that demonstrated lung stromal cells secrete high levels of HGF [18,272,273]. To investigate functional effects of HGF-MET signaling in lung stroma, we used WI-
38 lung fibroblast cells to determine to what extent they impact colonization of
TNBC cells. First, we validated that WI-38 cells activate MET receptors on TNBC cells. MDA-MB-231 and SUM159 cells stimulated with conditioned medium from
WI-38 cultures had significant MET activation (Figure 4.11b). We ensured the specificity of this activity by analyzing the secretome of WI-38 cells and found that these cells secrete ~7.2 ng/ml of HGF (Figure 4.11c).
To study the role of lung fibroblasts in survival and proliferation of TNBC cells, we developed a 3D lung stromal environment that consisted of a methylcellulose hydrogel containing dispersed WI-38 cells to represent lung stroma and single
TNBC cells to represent metastatic cells. We used four different TNBC cell lines at
114 a 1:4 ratio to WI-38 cells to mimic the relatively small population of cancer cells compared to the resident stromal cells in metastatic sites. Results showed that WI-
38 cells significantly promoted colony formation of SUM159, Hs578T, and MDA-
MB-157 cells (Figure 4.11d). To validate that this response was indeed due to the
HGF-MET pathway, we treated the cultures with a MET inhibitor, crizotinib.
Disrupting the HGF-MET signaling significantly reduced the colony formation of
SUM159, Hs578T, and MDA-MB-157 TNBC cells (Figure 4.11d). We further validated the impact of HGF-MET on survival and colony formation in lung stroma by immunofluorescent staining of the TNBC colonies with a proliferative cell marker, Ki-67. We observed a significant increase in Ki-67+ cells in TNBC colonies co-cultured with WI-38 cells than in vehicle controls that lacked the WI-38 cells, which was reversed by blocking HGF-MET axis (Figure 4.11e). We note that WI-
38 cells did not increase colony formation of MDA-MB-231 cells (data not shown), and further investigation is required to explain this phenomenon. Collectively, these results demonstrate that lung fibroblasts provide a permissive niche for survival and metastatic colonization of TNBC cells via HGF-MET axis.
4.3 Discussion
Tumor stroma is a major driver of tumorigenic activities and drug resistance of cancer cells [53,55,274]. Disrupting tumor-stromal interactions provides therapeutic opportunities especially for cancers such as TNBC that have poor prognosis and very limited targeted therapy options [275]. To mechanistically study tumor-stromal interactions and develop new treatment strategies that target the 115 tumor stroma, various 3D tumor models that recapitulate the complexity of the native TME have been developed as a preclinical tool to monitor drug efficacy
[150,275]. As detailed in Chapter III, we developed a 3D organotypic breast tumor model and demonstrated that it reproduces physicochemical and certain biologic properties of native tumors [253]. Uniquely, this model offers a broad design space to adjust cell and matrix composition and allows high throughput drug testing to identify treatments that effectively block tumorigenic functions of cancer cells.
Here, we incorporated patient-derived CAFs in this model to establish the utility of the model to identify therapeutic opportunities against tumor-stromal interactions.
Our finding that patient-derived CAFs predominantly secrete soluble factor,
HGF, is consistent with studies that showed CAFs in breast tumors secrete significant amount of HGF, as high as 50 ng/ml [48,276]. Although several reports have shown that breast cancer cells can produce HGF to activate MET in an autocrine signaling manner [277,278], we did not detect HGF in the supernatant of
TNBC cells. Furthermore, that all six TNBC cell lines that we used had endogenous
MET expression, in agreement with data from human breast tumors showing significantly higher MET gene expression in basal and claudin-low subtypes compared to other breast cancers (Figure 4.4). Other studies have shown that
MET is frequently overexpressed in malignant breast tissues [279] and TNBC tumors [280,281], and its expression is a strong independent prognostic factor in breast carcinoma [282–284]. Consistent with the role of CAFs in promoting pro- metastatic functions of breast cancer cells [48,53,285,286], CAFs from patients
116 significantly promoted proliferation, matrix invasion, and EMT of TNBC cells in our organotypic tumor model. These data suggests that paracrine signaling between
TNBC and patient-derived CAFs promote pro-metastatic functions in TNBC cells.
In addition to the chemical cues, TNBC cells also respond to the mechanical cues present within the TME. CAFs and cancer cells communicate through the ECM mediated by integrins and focal adhesion kinases (FAKs), promoting cell survival, migration, and invasion [287]. FAK associated with CAFs regulates breast cancer cell metabolism with low expression of FAK in CAFs associated with poor survival in human breast cancers [288].
Our analysis of RTK activities of TNBC cells showed high basal EGFR levels, consistent with EGFR overexpression in up to 78% of TNBC tumors
[279,289–292]. But in the presence of CAFs, the HGF-MET axis was the dominant mechanism of oncogenic signaling. This finding also highlights a potential reason for the failure of anti-EGFR monotherapies in TNBC [293], consistent with our result that showed that unlike EGFR inhibitors were ineffective against TNBC cell invasiveness, whereas crizotinib alone significantly reduced matrix invasion of
TNBC cells (Figure 4.12), indicating limitations of targeting only genetic aberrations of cancer cells and the importance of tumor-stromal interactions as a therapeutic target for TNBC.
Leveraging our model that allows mechanistic studies to guide treatment strategies, we investigated both single-agent and combination treatments against
HGF-MET and its downstream pathways. As expected, MDA-MB-231 cells
117 developed resistance to inhibition of MAPK pathway during long-term, cyclic treatments, highlighting the need for simultaneous blocking of more than a single signaling molecule. As a proof-of-concept, we designed and performed combination drug treatments that suppressed matrix invasion, proliferation, and
EMT of TNBC cells. The combinations targeted both the TNBC-CAFs interaction and an intrinsic active signaling in TNBC cells, i.e, crizotinib/trametinib for MDA-
MB-231 cells and crizotinib/dactolisib for SUM159 cells. This strategy is critical because targeting only the TNBC-CAFs interaction or genetic aberrations of TNBC cells proved insufficient. For example, using a MAPK inhibitor (trametinib) alone against MDA-MB-231 cells led to drug resistance, whereas targeting only the
CAFs-mediated MET activation using a MET inhibitor (crizotinib) alone did not block activities of ERK, Akt, and STAT3 in TNBC cells. We note that crizotinib alone was very effective in downregulating oncogenic pathways in TNBC cells with a brief 10 min treatment, but this was reversed over a 5-day treatment, indicating that TNBC cells become resistant to single-agent treatment of crizotinib as well.
The combination treatment successfully blocked activities of these pathways.
In addition to the role of HGF-MET in primary tumor progression, we showed that this signaling promoted colony forming capability of TNBC cells in a lung stromal environment. This is consistent with the role of stromal fibroblasts in promoting metastatic colonization of breast cancer cells in lung [263,294,295].
Other studies show that HGF significantly increased lung metastasis of breast cancer cells in mouse model studies [296,297]. Interestingly, disrupting HGF-MET
118 using crizotinib significantly reduced colony formation of TNBC cells in our model of lung stroma, suggesting that disrupting the HGF-MET paracrine signaling is a potential strategy to inhibit metastatic colonization of TNBC cells in the lungs.
Overall, this study underscores the value of our organotypic model to mechanistically study interactions of TNBC and stromal cells and identify key actionable targets. The model also allowed us to develop effective combination treatments that suppressed pro-metastatic processes such as matrix invasion, proliferation, and EMT of TNBC cells. This approach provides a potential drug discovery platform.
4.4 Summary
We demonstrated tumor-stromal interactions between TNBC and CAF cells in a physiologically relevant organotypic culture. We identified a specific soluble factor, HGF, secreted by CAFs that promotes pro-metastatic activities of TNBC cells. We showed the feasibility of targeting tumor stroma using design-driven drug combinations to significantly inhibit invasiveness, proliferation, and EMT of TNBC cells in a primary tumor environment and metastatic colonization of TNBC cells in a metastatic niche. Overall, this study offers a tool to study tumor-stromal interactions and develop novel treatments and strategies.
119
Table 4.1: List and sequences of primers of the genes analyzed for EMT of TNBC cells.
Primer Sequence Length
SNAIL F 5'-GGCTGCTACAAGGCCAT-3' 17
SNAIL R 5'-GCACTGGTACTTCTTGACATCT-3' 22
SLUG F 5'-AGGACACATTAGAACTCACACG-3' 22
SLUG R 5'-CAGATGAGCCCTCAGATTTGAC-3' 22
E47 F 5'-GCACAGACAAGGAGCTCAG-3' 19
E47 R 5'-GTCCTCAAGACCTGAACCTC-3' 20
ZEB1 F 5'-GGCATACACCTACTCAACTACG-3' 22
ZEB1 R 5'-CCTTCTGAGCTAGTATCTTGTCTTTC-3' 26
ZEB2 F 5'-CCTTTTTCTCCCCCACACTT-3' 20
ZEB2 R 5'-GATCAGATGGCAGTTCGCAT-3' 20
FOXC2 F 5'-GCAAACTTTCCCCAACGTG-3' 19
FOXC2 R 5'-CGGCGTGGATCTGTAGG-3' 17
TWIST F 5'-ATGTCCGCGTCCCACTA-3' 17
TWIST R 5'-ACTGTCCATTTTCTCCTTCTCTG-3' 23
120
Figure 4.1: HGF-MET signaling pathway. The binding of hepatocyte growth factor (HGF), the ligand of MET tyrosine kinase receptor (RTK), induces receptor dimerization and phosphorylation of multifunctional docking site in the kinase domain of the receptors. This activates many downstream signaling pathways including MAPK, PI3K/Akt, and STAT3.
121
Figure 4.2: CAFs-TNBC cells signal through HGF-MET pathway. (a) Phospho- RTK arrays of MDA-MB-231 and SUM159 cells stimulated with conditioned media of CAF-1 and CAF-2 monocultures. (b) Pixel densities representing activities of EGFR and MET from treatments normalized with the respective vehicle control and represented as activity fold change. Error bars represent standard errors from a mean value. Data represent three separate experiments. (c) Western blot analysis of effect of secretome of CAF-1 and CAF-2 on MET activation in six TNBC cell lines. (d) Quantified levels of p-MET normalizing with t-MET in TNBC cells stimulated with conditioned media of CAF-1 and CAF-2 cultures and statistically compared with a control group for each cell line. Data represent two separate experiments. *p < 0.05 was calculated using two-tailed, unpaired t-tests. (e) The levels of HGF, EGF, FGF-1, FGF-2, and PDGF AA/AB in the conditioned medium derived from fibroblasts cultures were determined using bead based multiplex immunoassay. CAF-1 and CAF-2 secreted HGF at significantly higher levels than normal human mammary fibroblast (HMF) did. Other soluble factors were not detectable. Data represent three separate experiments. *p < 0.001.
122
Figure 4.3: CAFs activate multiple oncogenic pathways in TNBC cells. (a) Western blot analysis shows CAF-1 and CAF-2 activate pathways beyond those already active, i.e., Akt and STAT3 in MDA-MB-231 cells and ERK1/2 and STAT3 in SUM159. (b) A MET inhibitor (crizotinib) downregulated oncogenic signaling in MDA-MB-231 and SUM159 cell lines. The graph shows quantified phospho- protein levels after normalizing with total protein levels in TNBC cells when stimulated with conditioned medium from CAF-1 with or without MET inhibition (0.5 µM and 1 µM crizotinib). Statistical comparison was done between TNBC cells stimulated with CAF-1 conditioned medium and the stimulated TNBC cells under crizotinib treatment. Data represent two separate experiments. *p < 0.05.
123
Figure 4.4: Met activity in EGFR+ TNBC. (a) Oncoprint of breast cancers (TCGA) shows higher EGFR and MET gene alterations among basal and claudin-low subtypes. (b) Analysis of METABRIC breast cancer database (TCGA) shows that MET and EGFR expression is significantly higher in subtypes representing TNBC disease. (c) MET expression strongly correlates with EGFR expression both at a gene level (METABRIC breast cancer database TCGA) and at a protein level (Firehose Legacy breast cancer database, TCGA). (d) Analysis of functional proteomics data of tumors from patients with invasive breast cancer derived from TCPA database shows a strong correlation between active p-MET and p-EGFR. *p < 0.05.
124
Figure 4.5: CAFs promote ECM invasion and proliferation of TNBC cells in a 3D organotypic tumor model. (a) A two-step micropatterning approach to create a tumor model that consists of a TNBC cell mass within a stroma composed of ECM and dispersed CAFs. Confocal reconstruction of tumor model (blue: TNBC mass; green: CAFs; collagen is not shown). (b) Confocal images of TNBC cells on day 5 of culture show matrix invasion of MDA-MB-231 and SUM159 cells promoted by CAFs, but not by HMFs. (c) Matrix invasion of MDA-MB-231 and SUM159 cells when cultured with CAF-1 and CAF-2 and normalized to the respective TNBC cells when cultured with HMFs. (d) A typical flow cytometry result from organotypic tumor models with HMF, CAF-1, or CAF-2, and TNBC cells. (e) Quantified flow cytometry results with absolute counts of MDA-MB-231 and SUM159 cells. * p<0.05.
125
Figure 4.6: CAFs promote EMT of TNBC cells. (a) CAFs promote a mesenchymal morphology in TNBC cells in organotypic models. (b,c) CAFs increase expression of EMT markers in TNBC cells both at gene and protein levels (*p<0.05 compared to vehicle control (+HMF)).
126
Figure 4.7: Screening of kinase inhibitors against TNBC spheroids. Dose−responses of (a) MDA-MB-231 and (b) SUM159 spheroids to inhibitors of MAPK pathway and PI3K pathway along with the list of molecular inhibitors used, their targets, and LD50 values against MDA-MB-231 and SUM159 spheroids. The ‘–‘ symbol indicates an LD50 value could not be obtained.
127
Figure 4.8: Cyclic drug treatment and recovery of MDA-B-231:CAF-1 co-culture spheroid. (a) The co-culture spheroids were cyclically treated with an inhibitor of MEK (5 nM trametinib). (b) Kinetics of spheroids growth during the cyclic treatment and recovery regimen. Each data point in the line graph is an average of eight replicates. (c) Percentage viability of TNBC cells measured from fluorescence intensity of endogenous GFP signal of TNBC cells (n=8). (d) Growth rate (kc) of co-culture spheroids during four treatment rounds with trametinib (n=8); *p<0.01. Error bars in panels (b-d) represent the standard errors from a mean value.
128
Figure 4.9: Targeting CAFs-mediated MDA-MB-231 pro-metastatic functions. (a) A typical matrix-format dose-dependent combination treatment with crizotinib (MET inhibitor) and trametinib (MEK1/2 inhibitor) to block matrix invasion of MDA- MB-231 cells. A combination concentration (red box) that significantly blocks invasiveness with a high synergy (CI<0.5) is shown. (b) Activity levels of ERK, Akt, and STAT3 in MDA-MB-231 organotypic cultures following combination treatments with 1 µM crizotinib and 5 nM trametinib. (c) Quantified levels of p-ERK/t-ERK, p- Akt/t-Akt, and p-STAT3/t-STAT3 followed by single-agent or combination treatments; *p<0.05. Data represent two separate experiments. (d) Quantified flow cytometry results with absolute counts of MDA-MB-231 from free-floating spheroids (MDA-MB-231 only) and organotypic cultures containing CAF-1 (control), single-agent treatments (+1 µM crizotinib or +5 nM trametinib) and combination treatment (+crizotinib+trametinib); * p<0.05 and n=5. (e) mRNA fold change values of seven EMT transcription factors before and after single-agent and combination treatments. Two-tailed, unpaired t-test was used to calculate the p-values; * p<0.05.
129
Figure 4.10: Targeting CAFs-mediated SUM159 pro-metastatic functions. (a) A typical matrix-format dose-dependent combination treatment with crizotinib (MET inhibitor) and dactolisib (PI3K inhibitor) to block matrix invasion of SUM159 cells. A combination concentration (red box) that significantly blocks invasiveness with a high synergy (CI<0.5) is shown. (b) Activity levels of ERK, Akt, and STAT3 in SUM159 organotypic cultures following combination treatments with 1 µM crizotinib and 1 µM dactolisib. (c) Quantified levels of p-ERK/t-ERK, p-Akt/t-Akt, and p-STAT3/t-STAT3 followed by single-agent or combination treatments; *p<0.05. Data represent two separate experiments. (d) Quantified flow cytometry results with absolute counts of SUM159 from free-floating spheroids (SUM159 only) and organotypic cultures containing CAF-1 (control), single-agent treatments (+1 µM crizotinib or +1 µM dactolisib) and combination treatment (+crizotinib+dactolisib); * p<0.05 and n=5. (e) mRNA fold change values of seven EMT transcription factors before and after single-agent and combination treatments. Two-tailed, unpaired t-test was used to calculate the p-values; * p<0.05.
130
Figure 4.11: HGF-MET signaling effect in a lung stromal environment. (a) MET signaling is enriched in murine breast cancer with lung metastases (GSE138139). The GSEA enrichment chart of MET signature genes in a murine model with lung metastases compared to parental TNBC cells. (b) Normal lung fibroblasts (WI-38) activated MET in TNBC cells. (c) WI-38 cells produced approximately 7.2 ng/ml of HGF. (d) WI-38 cells promoted formation of TNBC colonies in the lung stromal environment, and crizotinib significantly suppressed clone formation of SUM159, Hs578T, and MDA-MB-157 TNBC cells. (e) Immunofluorescence staining of Ki-67 (red) in SUM159, Hs578T, and MDA-MB-157 colonies with and without WI-38 cells or crizotinib. Green cells represent TNBC cells. Blot plots of normalized Ki-67+ cell count per spheroid area for three different conditions. The boxes represent the 25th and 75th percentiles with the median shown with a horizontal line inside each box. The mean is shown with a cross symbol. The whiskers represent the 10th and 90th percentile of the data. Two-tailed Mann-Whitney test was used to calculate p- value; *p<0.01 and n=7.
131
Figure 4.12: Confocal images of organotypic cultures of MDA-MB-231 and SUM159 spheroids on day 5 of culture with and without inhibitors (1 µM crizotinib, neratinib, or lapatinib). The graphs show normalized invasion area of MDA-MB- 231 and SUM159 cells with that of vehicle control; *p<0.01 and n=6.
132
CHAPTER V
CONCLUSIONS
We developed a 3D organotypic breast tumor model for mechanistic studies of tumor-stromal interactions and provided proof-of-concept studies of therapeutic approaches that target the tumor stroma. We first developed the organotypic culture using a cell and protein micropatterning in aqueous two-phase system
(ATPS). We then validated our 3D tumor model using a well-known biological signaling axis that involves a fibroblast-secreted chemokine CXCL12 and CXCR4 receptor on TNBC cells to recapitulate major hallmark of cancer, i.e., cancer cell invasion. Finally, we mechanistically studied interactions between patient-derived cancer-associated fibroblasts (CAFs) and triple negative breast cancer (TNBC) cells to facilitate design and testing of stroma-targeted treatments.
Major conclusions of this study are as follows:
(1) In Chapter II, we used a novel method of a cell and protein micropatterning approach using ATPS. We showed that partition of collagen in PEG-DEX ATPS is sensitive to polymer molecular weight. Using a PEG 35 kDa–DEX 500 kDa formulation, we showed partition of collagen to the bottom DEX phase of ATPS and employed this approach to conveniently develop 3D breast tumor models. The
3D tumor model was generated in two pipetting steps: first a DEX phase nanodrop
133 containing cancer cells was dispensed in the PEG phase to form a 3D mass
(spheroid). Next, a collagen-containing DEX microdrop was dispensed to combine with spheroid-containing nanodrop to form a collagen hydrogel that entrapped the spheroid. This approach conveniently produced collagen gels of desired concentrations to help reproduce mechanical properties of solid tumors.
(2) In Chapter III, we developed a breast tumor model by incorporating three main components of solid tumors: a mass of breast cancer cell, stromal fibroblasts, and extracellular matrix (ECM). The resulting product resembled the architecture of solid tumors in terms of spatial distribution of cells and ECM. To establish the validity of our model, we used TNBC with CXCL12-CXCR4 signaling as a disease model and showed that fibroblasts-secreted CXCL12 promoted ECM invasion of
CXCR4+ TNBC cells by activating oncogenic mitogen-activated protein kinase
(MAPK) pathway. The CXCL12 secreting fibroblasts remodeled the matrix by contracting the collagen matrix through RhoA/ROCK/myosin light chain-2 pathway. These results were consistent with prior literature on the role of this signaling axis on tumorigenic activities of cancer cells. We demonstrated the feasibility of blocking tumor-stromal interactions as a therapeutic strategy by suppressing CXCL12-induced ECM invasion of TNBC cells using a molecular inhibitor (AMD3100) that blocked this signaling axis.
(3) In Chapter IV, we established the utility of our organotypic tumor model to study dynamic interactions between patient-derived CAFs and TNBC cells. We showed that these interactions were predominantly through activation of MET receptor on
TNBC cells by CAFs-secreted HGF. Using our organotypic tumor model, we
134
showed that CAFs promote proliferation, invasiveness, and epithelial-to- mesenchymal transition (EMT) in TNBC cells by activating oncogenic pathways such as MAPK, PI3K/Akt, and STAT3. Importantly, we found that TNBC cells are resistant to single-agent treatment with a potent MEK inhibitor (trametinib), and that using a design-driven combination treatment was effective against pro- metastatic functions of TNBC cells. Using our organotypic culture, we performed matrix-style combination treatments and identified a pair of drug combinations that reduced proliferation, invasiveness, and EMT of TNBC cells. We also demonstrated that the HGF-MET axis is implicated in lung metastasis of TNBC and that blocking this signaling is a potential approach against primary TNBC tumors and metastases formation in the lung.
Altogether, this research addressed an unmet need for a scalable, physiologically relevant 3D tumor model that facilitates tumor-stromal interactions and testing of new treatments against tumor stroma. The modularity of this model allows convenient adjustments to the composition of the ECM and the stromal cells such as immune and endothelial cells to explore other avenues of tumor-stromal interactions. Ultimately, this study shed light on the feasibility of using the tumor microenvironment as a therapy target especially for cancers such as TNBC that have very limited targeted therapy options.
135 CHAPTER VI
FUTURE WORK
The studies presented here demonstrate the potential of our biomimetic breast tumor model for future investigations of various other tumor-stromal interactions that are currently difficult to accurately capture with existing in vitro and animal models. While our tumor model consists of three key components of the tumor microenvironment (TME), i.e., cancer cells, fibroblasts, and collagen, our aqueous two-phase system (ATPS) micropatterning technology enables modular addition of different stromal cells and extracellular matrix proteins to this 3D tumor model. The ability to study interactions of cancer cells and various stromal cells using our organotypic tumor model will enable a mechanistic understanding of the
TME to identify prominent molecular drivers of the disease and allow testing of arrays of drugs to identify effective therapeutics.
Following are several potential topics of future studies.
(1) Extracellular matrix (ECM) components: Throughout our studies, we used breast cancer cells and cancer-associated fibroblasts embedded in the ECM to perform phenotypic and mechanistic studies of tumor-stromal interactions.
However, tumors reside in a more complex microenvironment that also contains other cellular and non-cellular components such as immune cells, endothelial cells, and other ECM proteins [20]. Interactions of cancer cells with these components
136 of the TME regulate tumor growth, disease progression, and drug resistance, which we reviewed in Chapter I. Particularly, ECM contains various proteins such as type I collagen, type IV collagen, laminin, hyaluronic acid, fibronectin, and proteoglycans [298]. We used type I collagen as an ECM protein in our model because it is the dominant component of breast tissue ECM. However, interactions of other ECM proteins with cancer cells have significant effects on tumorigenic function of cancer cells [299]. Our ATPS micropatterning technology allows straightforward incorporation of various ECM proteins into the tumor model for mechanistic studies of effect of tumor-ECM signaling. Understanding molecular mechanisms of ECM-mediated tumor-stromal interactions may lead to discovery of new treatments to block malignancy of cancer cells.
(2) Long-term culture of organotypic tumor model: Throughout this project, we used organotypic cultures with single-agent and combination drug treatments that lasted for only 5 days. We assessed the efficacy of the drugs based primarily on inhibition of invasiveness and proliferation of cancer cells within this period.
However, despite an initial efficacy of single-agent treatments, cancer cells often develop resistance to a given drug over a much longer time [115]. In Chapter IV, we showed a proof-of-concept study of adaptive drug resistance using a co-culture spheroid model of breast cancer cells and CAFs under a long-term, cyclic treatment and recovery regimen that mimics cyclic chemotherapy of patients.
Future direction includes use of this regimen with the organotypic cultures to assess the long-term efficacy of different treatments against matrix invasion and proliferation of cancer cells and activities of oncogenic signaling pathways. A
137
potential impediment will be quantitative studies of invasion and proliferation based on endogenous GFP signal of cancer cells because fibroblast cells constantly remodel and shrink the collagen matrix. A possible solution is to use multi-photon microscopy which provides a higher penetration depth of up to 1.6 mm, to image dense organotypic cultures [300,301].
(3) Validation in xenografts. It is well recognized that accurate modeling of complexities and heterogeneity of tumors in preclinical studies is a challenge
[11,249]. Assessing drug efficacy preclinically using models that lack multiple components of the TME, and thus missing out the tumor-stromal interactions, is one of the major reasons of high attrition rate of cancer drugs. While animal models are more physiologic than in vitro models and contains vasculature, they lack human stroma and immune system. Therefore, it is important for preclinical studies to utilize both in vitro 3D tumor models and tumor xenografts to both mechanistically study effectiveness of different treatment regimens. As such, our future studies should use PDX models to validate the performance of predicting the efficacy of drugs in the organotypic cultures. Immediate questions to answer are whether reduction in invasiveness and inhibition of proliferation of cancer cells in our tumor model translates respectively into reduction in local invasion and metastasis, and shrinkage of tumors in the PDX model. Comparison of results from these models will ensure identifying the most effective drug combinations prior to clinical trials.
(4) Fully humanized tumor model: In our study, we incorporated TNBC cell lines, rat tail type I collagen, and patient-derived CAFs in our organotypic tumor
138 model to study the tumor-stromal interactions. While we showed the feasibility of working with primary cells, the model was not fully humanized. In future studies, incorporating human-derived ECM proteins and patient-derived cancer and stromal cells will lead to a fully humanized tool for mechanistic studies and phenotypic screening of drugs. However, potential difficulties will be maintaining primary tumor cells in vitro for long-term cultures and low cell population yield. A potential solution is to use the technique of conditional reprogramming of primary cells that allow expanding the cells in sufficient numbers and for many passages
[302]. Conditionally reprogrammed cells have been shown to maintain properties of their native tumors and allow identifying effective therapies for specific patients
[303].
(5) Personalized medicine: Triple negative breast cancer (TNBC) is a heterogeneous disease with several molecular subtypes [15]. The heterogeneity of tumors results in significant differences in therapy responses among patients of the same cancer type [245]. The field of cancer medicine is shifting toward personalized medicine that utilizes patient-specific tumor models containing patient-derived cancer and stromal cells to identify effective therapies for specific patients [303]. Analytical techniques such as next generation sequencing will help identify major therapeutic targets which could be therapeutically targeted in the tumor model to identify the most effective treatments for individual patients. Our organotypic tumor model is a versatile tool and is compatible with culturing of patient-derived cells. Future studies should investigate the feasibility of generating patient-specific tumor models to test arrays of drugs and their combinations based
139 on molecular drivers of the disease in individual patients. Our organotypic tumor model has the potential for preclinical studies of personalized cancer medicine.
140
BIBLIOGRAPHY
[1] R.L. Siegel, K.D. Miller, A. Jemal, Cancer statistics, 2019, CA. Cancer J. Clin. 69 (2019) 7–34. https://doi.org/10.3322/caac.21551.
[2] F. Bray, J. Ferlay, I. Soerjomataram, R.L. Siegel, L.A. Torre, A. Jemal, Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries, CA. Cancer J. Clin. (2018). https://doi.org/10.3322/caac.21492.
[3] C.E. DeSantis, J. Ma, M.M. Gaudet, L.A. Newman, K.D. Miller, A. Goding Sauer, A. Jemal, R.L. Siegel, Breast cancer statistics, 2019, CA. Cancer J. Clin. 69 (2019) 438–451. https://doi.org/10.3322/caac.21583.
[4] I. Jatoi, A.B. Miller, Why is breast-cancer mortality declining?, Lancet Oncol. 4 (2003) 251–254. https://doi.org/10.1016/S1470-2045(03)01037-4.
[5] Y. Gao, A.E. Teschendorff, Epigenetic and genetic deregulation in cancer target distinct signaling pathway domains, Nucleic Acids Res. 45 (2017) 583–596. https://doi.org/10.1093/nar/gkw1100.
[6] J.A. Joyce, Therapeutic targeting of the tumor microenvironment, Cancer Cell. (2005) 513–520. https://doi.org/10.1016/j.ccr.2005.05.024.
[7] K. Pietras, A. Östman, Hallmarks of cancer: Interactions with the tumor stroma, Exp. Cell Res. 316 (2010) 1324–1331. https://doi.org/10.1016/j.yexcr.2010.02.045.
[8] Y. Mao, E.T. Keller, D.H. Garfield, K. Shen, J. Wang, Stromal cells in tumor microenvironment and breast cancer, Cancer Metastasis Rev. 32 (2013) 303–315. https://doi.org/10.1007/s10555-012-9415-3.
[9] A. Bahrami, S.M. Hassanian, M. Khazaei, M. Hasanzadeh, S. Shahidsales, M. Maftouh, G.A. Ferns, A. Avan, The Therapeutic Potential of Targeting Tumor Microenvironment in Breast Cancer: Rational Strategies and Recent Progress, J. Cell. Biochem. 119 (2018) 111–122. https://doi.org/10.1002/jcb.26183.
[10] A.S. Cazet, M.N. Hui, B.L. Elsworth, S.Z. Wu, D. Roden, C.L. Chan, J.N.
141
Skhinas, R. Collot, J. Yang, K. Harvey, M.Z. Johan, C. Cooper, R. Nair, D. Herrmann, A. McFarland, N. Deng, M. Ruiz-Borrego, F. Rojo, J.M. Trigo, S. Bezares, R. Caballero, E. Lim, P. Timpson, S. O’Toole, D.N. Watkins, T.R. Cox, M.S. Samuel, M. Martín, A. Swarbrick, Targeting stromal remodeling and cancer stem cell plasticity overcomes chemoresistance in triple negative breast cancer, Nat. Commun. 9 (2018). https://doi.org/10.1038/s41467-018-05220-6.
[11] M.R. Junttila, F.J. De Sauvage, Influence of tumour micro-environment heterogeneity on therapeutic response, Nature. 501 (2013) 346–354. https://doi.org/10.1038/nature12626.
[12] M. Kalimutho, K. Parsons, D. Mittal, J.A. López, S. Srihari, K.K. Khanna, Targeted Therapies for Triple-Negative Breast Cancer: Combating a Stubborn Disease, Trends Pharmacol. Sci. 36 (2015) 822–846. https://doi.org/10.1016/j.tips.2015.08.009.
[13] P. Eroles, A. Bosch, J. Alejandro Pérez-Fidalgo, A. Lluch, Molecular biology in breast cancer: Intrinsic subtypes and signaling pathways, Cancer Treat. Rev. 38 (2012) 698–707. https://doi.org/10.1016/j.ctrv.2011.11.005.
[14] V. Eusebi, Classifications and prognosis of breast cancer: From morphology to molecular taxonomy, Breast J. 16 (2010) S15–S16. https://doi.org/10.1111/j.1524-4741.2010.00995.x.
[15] G. Bianchini, J.M. Balko, I.A. Mayer, M.E. Sanders, L. Gianni, Triple- negative breast cancer: Challenges and opportunities of a heterogeneous disease, Nat. Rev. Clin. Oncol. 13 (2016) 674–690. https://doi.org/10.1038/nrclinonc.2016.66.
[16] F. Kassam, K. Enright, R. Dent, G. Dranitsaris, J. Myers, C. Flynn, M. Fralick, R. Kumar, M. Clemons, Survival outcomes for patients with metastatic triple-negative breast cancer: Implications for clinical practice and trial design, Clin. Breast Cancer. 9 (2009) 29–33. https://doi.org/10.3816/CBC.2009.n.005.
[17] L.A. Carey, C.M. Perou, C.A. Livasy, L.G. Dressler, D. Cowan, K. Conway, G. Karaca, M.A. Troester, K.T. Chiu, S. Edmiston, S.L. Deming, J. Geradts, M.C.U. Cheang, T.O. Nielsen, P.G. Moorman, H.S. Earp, R.C. Millikan, Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study, J. Am. Med. Assoc. 295 (2006) 2492–2502. https://doi.org/10.1001/jama.295.21.2492.
[18] T. Straussman, R. Morikawa, K. Shee, M. Barzily-Rokni, Z.R. Qian, J. Du, A. Davis, M.M. Mongare, J. Gould, D.T. Frederick, Z.A. Cooper, Tumour micro-environment elicits innate resistance to RAF inhibitors through HGF 142
secretion, Nature. 487 (2012) 500–504. https://doi.org/10.7554/eLife.04034.
[19] A.M. Moorman, R. Vink, H.J. Heijmans, J. Van Der Palen, E.A. Kouwenhoven, The prognostic value of tumour-stroma ratio in triple- negative breast cancer, Eur. J. Surg. Oncol. 38 (2012) 307–313. https://doi.org/10.1016/j.ejso.2012.01.002.
[20] F.R. Balkwill, M. Capasso, T. Hagemann, The tumor microenvironment at a glance, J. Cell Sci. 125 (2012) 5591–5596. https://doi.org/10.1242/jcs.116392.
[21] S. Winslow, K. Leandersson, A. Edsjö, C. Larsson, Prognostic stromal gene signatures in breast cancer, Breast Cancer Res. 17 (2015) 1–13. https://doi.org/10.1186/s13058-015-0530-2.
[22] S.D. Soysal, A. Tzankov, S.E. Muenst, Role of the Tumor Microenvironment in Breast Cancer, Pathobiology. 82 (2015) 142–152. https://doi.org/10.1159/000430499.
[23] R.E. Coleman, W. Gregory, H. Marshall, C. Wilson, I. Holen, The metastatic microenvironment of breast cancer: Clinical implications, Breast. 22 (2013). https://doi.org/10.1016/j.breast.2013.07.010.
[24] L. Jin, B. Han, E. Siegel, Y. Cui, A. Giuliano, X. Cui, Breast cancer lung metastasis: Molecular biology and therapeutic implications, Cancer Biol. Ther. 19 (2018) 858–868. https://doi.org/10.1080/15384047.2018.1456599.
[25] H. Kennecke, R. Yerushalmi, R. Woods, M.C.U. Cheang, D. Voduc, C.H. Speers, T.O. Nielsen, K. Gelmon, Metastatic behavior of breast cancer subtypes, J. Clin. Oncol. 28 (2010) 3271–3277. https://doi.org/10.1200/JCO.2009.25.9820.
[26] S.M. Pupa, S. Ménard, S. Forti, E. Tagliabue, New insights into the role of extracellular matrix during tumor onset and progression, J. Cell. Physiol. 192 (2002) 259–267. https://doi.org/10.1002/jcp.10142.
[27] M. Egeblad, M.G. Rasch, V.M. Weaver, Dynamic interplay between the collagen scaffold and tumor evolution, Curr. Opin. Cell Biol. 22 (2010) 697– 706. https://doi.org/10.1016/j.ceb.2010.08.015.
[28] M.J. Paszek, N. Zahir, K.R. Johnson, J.N. Lakins, G.I. Rozenberg, A. Gefen, C.A. Reinhart-King, S.S. Margulies, M. Dembo, D. Boettiger, D.A. Hammer, V.M. Weaver, Tensional homeostasis and the malignant phenotype, Cancer Cell. 8 (2005) 241–254. https://doi.org/10.1016/j.ccr.2005.08.010. 143
[29] D.T. Butcher, T. Alliston, V.M. Weaver, A tense situation: Forcing tumour progression, Nat. Rev. Cancer. 9 (2009) 108–122. https://doi.org/10.1038/nrc2544.
[30] P.P. Provenzano, D.R. Inman, K.W. Eliceiri, P.J. Keely, Matrix density- induced mechanoregulation of breast cell phenotype, signaling and gene expression through a FAK-ERK linkage, Oncogene. 28 (2009) 4326–4343. https://doi.org/10.1038/onc.2009.299.
[31] M.J. Paszek, V.M. Weaver, The tension mounts: Mechanics meets morphogenesis and malignancy, J. Mammary Gland Biol. Neoplasia. 9 (2004) 325–342. https://doi.org/10.1007/s10911-004-1404-x.
[32] M. Fang, J. Yuan, C. Peng, Y. Li, Collagen as a double-edged sword in tumor progression, Tumor Biol. 35 (2014) 2871–2882. https://doi.org/10.1007/s13277-013-1511-7.
[33] M. Hayashi, Y. Yamamoto, M. Ibusuki, S. Fujiwara, S. Yamamoto, S. Tomita, M. Nakano, K. Murakami, K.I. Iyama, H. Iwase, Evaluation of tumor stiffness by elastography is predictive for pathologic complete response to neoadjuvant chemotherapy in patients with breast cancer, Ann. Surg. Oncol. 19 (2012) 3042–3049. https://doi.org/10.1245/s10434-012-2343-1.
[34] N.F. Boyd, Q. Li, O. Melnichouk, E. Huszti, L.J. Martin, A. Gunasekara, G. Mawdsley, M.J. Yaffe, S. Minkin, Evidence that breast tissue stiffness is associated with risk of breast cancer, PLoS One. 9 (2014) e100937. https://doi.org/10.1371/journal.pone.0100937.
[35] F. Grinnell, R.E. Billingham, L. Burgess, Distribution of fibronectin during wound healing in vivo, J. Invest. Dermatol. 76 (1981) 181–189. https://doi.org/10.1111/1523-1747.ep12525694.
[36] M. Simian, Y. Hirai, M. Navre, Z. Werb, A. Lochter, M.J. Bissell, The interplay of matrix metalloproteinases, morphogens and growth factors is necessary for branching of mammary epithelial cells, Development. 128 (2001) 3117–3131. https://dev.biologists.org/content/128/16/3117.short.
[37] J.T. Oxford, J.C. Reeck, M.J. Hardy, Extracellular matrix in development and disease, Int. J. Mol. Sci. 20 (2019). https://doi.org/10.3390/ijms20010205.
[38] P. Garin-Chesa, L.J. Old, W.J. Rettig, Cell surface glycoprotein of reactive stromal fibroblasts as a potential antibody target in human epithelial cancers, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 7235–7239. https://doi.org/10.1073/pnas.87.18.7235.
144
[39] B. Li, J.H.C. Wang, Fibroblasts and myofibroblasts in wound healing: Force generation and measurement, J. Tissue Viability. 20 (2011) 108–120. https://doi.org/10.1016/j.jtv.2009.11.004.
[40] F. Xing, J. Saidou, K. Watabe, Cancer associated fibroblasts (CAFs) in tumor microenvironment., Front. Biosci. 15 (2010) 166–179. https://doi.org/10.2741/3613.
[41] J.S. Flier, L.H. Underhill, H.F. Dvorak, Tumors: Wounds That Do Not Heal, N. Engl. J. Med. 315 (1986) 1650–1659. https://doi.org/10.1056/NEJM198612253152606.
[42] L.E. Littlepage, M. Egeblad, Z. Werb, Coevolution of cancer and stromal cellular responses, Cancer Cell. 7 (2005) 499–500. https://doi.org/10.1016/j.ccr.2005.05.019.
[43] N.A. Bhowmick, E.G. Neilson, H.L. Moses, Stromal fibroblasts in cancer initiation and progression, Nature. 432 (2004) 332–337. https://doi.org/10.1038/nature03096.
[44] K. Amornsupak, P. Jamjuntra, M. Warnnissorn, P. O Charoenrat, D. Sanguanraksa, P. Thuwajit, S.A. Eccles, C. Thuwajit, P. O-Charoenrat, D. Sa-nguanraksa, P. Thuwajit, S.A. Eccles, C. Thuwajit, High ASMA + Fibroblasts and Low Cytoplasmic HMGB1 + Breast Cancer Cells Predict Poor Prognosis, Clin. Breast Cancer. 17 (2017) 441–452. https://doi.org/10.1016/j.clbc.2017.04.007.
[45] G. Ishii, A. Ochiai, S. Neri, Phenotypic and functional heterogeneity of cancer-associated fibroblast within the tumor microenvironment, Adv. Drug Deliv. Rev. 99 (2016) 186–196. https://doi.org/10.1016/j.addr.2015.07.007.
[46] L. Rønnov-Jessen, O.W. Petersen, V.E. Koteliansky, M.J. Bissell, The origin of the myofibroblasts in breast cancer: Recapitulation of tumor environment in culture unravels diversity and implicates converted fibroblasts and recruited smooth muscle cells, J. Clin. Invest. 95 (1995) 859–873. https://doi.org/10.1172/JCI117736.
[47] W.J. Chen, C.C. Ho, Y.L. Chang, H.Y. Chen, C.A. Lin, T.Y. Ling, S.L. Yu, S.S. Yuan, Y.J. Louisa Chen, C.Y. Lin, S.H. Pan, H.Y. Elizabeth Chou, G.C. Chang, W.C. Chu, Y.M. Lee, J.Y. Lee, P.J. Lee, K.C. Li, H.W. Chen, P.C. Yang, Cancer-associated fibroblasts regulate the plasticity of lung cancer stemness via paracrine signalling, Nat. Commun. 5 (2014). https://doi.org/10.1038/ncomms4472.
[48] S.W. Tyan, W.H. Kuo, C.K. Huang, C.C. Pan, J.Y. Shew, K.J. Chang, E.Y.H.P. Lee, W.H. Lee, Breast cancer cells induce cancer-associated 145
fibroblasts to secrete hepatocyte growth factor to enhance breast tumorigenesis, PLoS One. 6 (2011) e15313. https://doi.org/10.1371/journal.pone.0015313.
[49] X. Wu, X. Chen, Q. Zhou, P. Li, B. Yu, J. Li, Y. Qu, J. Yan, Y. Yu, M. Yan, Z. Zhu, B. Liu, L. Su, Hepatocyte growth factor activates tumor stromal fibroblasts to promote tumorigenesis in gastric cancer, Cancer Lett. 335 (2013) 128–135. https://doi.org/10.1016/j.canlet.2013.02.002.
[50] C.J. Powers, S.W. McLeskey, A. Wellstein, Fibroblast growth factors, their receptors and signaling, Endocr. Relat. Cancer. 7 (2000) 165–197. https://doi.org/10.1677/erc.0.0070165.
[51] X. Wu, P. Tao, Q. Zhou, J. Li, Z. Yu, X. Wang, J. Li, C. Li, M. Yan, Z. Zhu, B. Liu, L. Su, IL-6 secreted by cancer-associated fibroblasts promotes epithelial-mesenchymal transition and metastasis of gastric cancer via JAK2/STAT3 signaling pathway, Oncotarget. 8 (2017) 20741–20750. https://doi.org/10.18632/oncotarget.15119.
[52] X. Sun, Y. Mao, J. Wang, L. Zu, M. Hao, G. Cheng, Q. Qu, D. Cui, E.T. Keller, X. Chen, K. Shen, J. Wang, IL-6 secreted by cancer-associated fibroblasts induces tamoxifen resistance in luminal breast cancer, Oncogene. 33 (2014) 4450. https://doi.org/10.1038/onc.2014.224.
[53] A. Orimo, P.B. Gupta, D.C. Sgroi, F. Arenzana-Seisdedos, T. Delaunay, R. Naeem, V.J. Carey, A.L. Richardson, R.A. Weinberg, Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion, Cell. 121 (2005) 335–348. https://doi.org/10.1016/j.cell.2005.02.034.
[54] P. Ray, A.C. Stacer, J. Fenner, S.P. Cavnar, K. Meguiar, M. Brown, K.E. Luker, G.D. Luker, CXCL12-γ in primary tumors drives breast cancer metastasis, Oncogene. 34 (2014) 2043–2051. https://doi.org/10.1038/onc.2014.157.
[55] L. Tao, G. Huang, H. Song, Y. Chen, L. Chen, Cancer associated fibroblasts: An essential role in the tumor microenvironment, Oncol. Lett. 14 (2017) 2611–2620. https://doi.org/10.3892/ol.2017.6497.
[56] F. Al-Ejeh, W. Shi, M. Miranda, P.T. Simpson, A.C. Vargas, S. Song, A.P. Wiegmans, A. Swarbrick, A.L. Welm, M.P. Brown, G. Chenevix-Trench, S.R. Lakhani, K.K. Khanna, Treatment of triple-negative breast cancer using anti-EGFR-directed radioimmunotherapy combined with radiosensitizing chemotherapy and PARP inhibitor, J. Nucl. Med. 54 (2013) 913–921. https://doi.org/10.2967/jnumed.112.111534.
146
[57] N. V. Sergina, M. Rausch, D. Wang, J. Blair, B. Hann, K.M. Shokat, M.M. Moasser, Escape from HER-family tyrosine kinase inhibitor therapy by the kinase-inactive HER3, Nature. 445 (2007) 437–441. https://doi.org/10.1038/nature05474.
[58] Y. Du, H. Yamaguchi, Y. Wei, J.L. Hsu, H.L. Wang, Y.H. Hsu, W.C. Lin, W.H. Yu, P.G. Leonard, G.R. Lee, M.K. Chen, K. Nakai, M.C. Hsu, C. Te Chen, Y. Sun, Y. Wu, W.C. Chang, W.C. Huang, C.L. Liu, Y.C. Chang, C.H. Chen, M. Park, P. Jones, G.N. Hortobagyi, M.C. Hung, Blocking c- Met-mediated PARP1 phosphorylation enhances anti-tumor effects of PARP inhibitors, Nat. Med. 22 (2016) 194–201. https://doi.org/10.1038/nm.4032.
[59] R. Sharpe, A. Pearson, M.T. Herrera-Abreu, D. Johnson, A. Mackay, J.C. Welti, R. Natrajan, A.R. Reynolds, J.S. Reis-Filho, A. Ashworth, N.C. Turner, FGFR signaling promotes the growth of triple-negative and basal- like breast cancer cell lines both in vitro and in vivo, Clin. Cancer Res. 17 (2011) 5275–5286. https://doi.org/10.1158/1078-0432.CCR-10-2727.
[60] C.L. Addison, G.R. Pond, B. Cochrane, H. Zhao, S.K. Chia, M.N. Levine, M. Clemons, Correlation of baseline biomarkers with clinical outcomes and response to fulvestrant with vandetanib or placebo in patients with bone predominant metastatic breast cancer: An OCOG ZAMBONEY sub-study, J. Bone Oncol. 4 (2015) 47–53. https://doi.org/10.1016/j.jbo.2015.04.001.
[61] S. Liang, X. Peng, X. Li, P. Yang, L. Xie, Y. Li, C. Du, G. Zhang, Silencing of CXCR4 sensitizes triple-negative breast cancer cells to cisplatin, Oncotarget. 6 (2015) 1020–1030. https://doi.org/10.18632/oncotarget.2741.
[62] P. Ray, S.A. Lewin, L.A. Mihalko, B.T. Schmidt, K.E. Luker, G.D. Luker, Noninvasive imaging reveals inhibition of ovarian cancer by targeting CXCL12-CXCR4, Neoplasia. 13 (2011) 1152–1161. https://doi.org/10.1593/neo.111076.
[63] K.E. De Visser, A. Eichten, L.M. Coussens, Paradoxical roles of the immune system during cancer development, Nat. Rev. Cancer. 6 (2006) 24–37. https://doi.org/10.1038/nrc1782.
[64] R.D. Leek, C.E. Lewis, R. Whitehouse, M. Greenall, J. Clarke, A.L. Harris, Association of macrophage infiltration with angiogenesis and prognosis in invasive breast carcinoma, Cancer Res. 56 (1996) 4625–4629. https://pubmed.ncbi.nlm.nih.gov/8840975/.
[65] N. Linde, M. Casanova-Acebes, M.S. Sosa, A. Mortha, A. Rahman, E. Farias, K. Harper, E. Tardio, I. Reyes Torres, J. Jones, J. Condeelis, M. 147
Merad, J.A. Aguirre-Ghiso, Macrophages orchestrate breast cancer early dissemination and metastasis, Nat. Commun. 9 (2018). https://doi.org/10.1038/s41467-017-02481-5.
[66] J. Condeelis, J.W. Pollard, Macrophages: Obligate partners for tumor cell migration, invasion, and metastasis, Cell. 124 (2006) 263–266. https://doi.org/10.1016/j.cell.2006.01.007.
[67] M. Kraman, P.J. Bambrough, J.N. Arnold, E.W. Roberts, L. Magiera, J.O. Jones, A. Gopinathan, D.A. Tuveson, D.T. Fearon, Suppression of antitumor immunity by stromal cells expressing fibroblast activation protein- α, Science (80-. ). 330 (2010) 827–830. https://doi.org/10.1126/science.1195300.
[68] P.B. Olkhanud, B. Damdinsuren, M. Bodogai, R.E. Gress, R. Sen, K. Wejksza, E. Malchinkhuu, R.P. Wersto, A. Biragyn, Tumor-evoked regulatory B cells promote breast cancer metastasis by converting resting CD4+ T cells to T-regulatory cells, Cancer Res. 71 (2011) 3505–3515. https://doi.org/10.1158/0008-5472.CAN-10-4316.
[69] K.M. Bussard, L. Mutkus, K. Stumpf, C. Gomez-Manzano, F.C. Marini, Tumor-associated stromal cells as key contributors to the tumor microenvironment, Breast Cancer Res. 18 (2016) 84. https://doi.org/10.1186/s13058-016-0740-2.
[70] K. Hida, Y. Hida, D.N. Amin, A.F. Flint, D. Panigrahy, C.C. Morton, M. Klagsbrun, Tumor-associated endothelial cells with cytogenetic abnormalities, Cancer Res. 64 (2004) 8249–8255. https://doi.org/10.1158/0008-5472.CAN-04-1567.
[71] L. V. Beerepoot, N. Mehra, J.S.P. Vermaat, B.A. Zonnenberg, M.F.G.B. Gebbink, E.E. Voest, Increased levels of viable circulating endothelial cells are an indicator of progressive disease in cancer patients, Ann. Oncol. 15 (2004) 139–145. https://doi.org/10.1093/annonc/mdh017.
[72] A. Yadav, B. Kumar, J.-G. Yu, M. Old, T.N. Teknos, P. Kumar, Tumor- Associated Endothelial Cells Promote Tumor Metastasis by Chaperoning Circulating Tumor Cells and Protecting Them from Anoikis., PLoS One. 10 (2015) e0141602. https://doi.org/10.1371/journal.pone.0141602.
[73] A.L. Ribeiro, O.K. Okamoto, Combined effects of pericytes in the tumor microenvironment, Stem Cells Int. 2015 (2015). https://doi.org/10.1155/2015/868475.
[74] K. Ochs, F. Sahm, C.A. Opitz, T. V. Lanz, I. Oezen, P.O. Couraud, A. von Deimling, W. Wick, M. Platten, Immature mesenchymal stem cell-like 148
pericytes as mediators of immunosuppression in human malignant glioma, J. Neuroimmunol. 265 (2013) 106–116. https://doi.org/10.1016/j.jneuroim.2013.09.011.
[75] A. Cretu, P.C. Brooks, Impact of the non-cellular tumor microenvironment on metastasis: Potential therapeutic and imaging opportunities, J. Cell. Physiol. 213 (2007) 391–402. https://doi.org/10.1002/jcp.21222.
[76] Y. Yuan, Y.C. Jiang, C.K. Sun, Q.M. Chen, Role of the tumor microenvironment in tumor progression and the clinical applications (Review), Oncol. Rep. 35 (2016) 2499–2515. https://doi.org/10.3892/or.2016.4660.
[77] N.M. Anderson, M.C. Simon, The tumor microenvironment, Curr. Biol. 30 (2020) R921–R925. https://doi.org/10.1016/j.cub.2020.06.081.
[78] T.R. Wilson, J. Fridlyand, Y. Yan, E. Penuel, L. Burton, E. Chan, J. Peng, E. Lin, Y. Wang, J. Sosman, A. Ribas, J. Li, J. Moffat, D.P. Sutherlin, H. Koeppen, M. Merchant, R. Neve, J. Settleman, Widespread potential for growth-factor-driven resistance to anticancer kinase inhibitors, Nature. 487 (2012) 505–509. https://doi.org/10.1038/nature11249.
[79] Q. Yu, Y. Xue, J. Liu, Z. Xi, Z. Li, Y. Liu, Fibronectin promotes the malignancy of glioma stem-like cells via modulation of cell adhesion, differentiation, proliferation and chemoresistance, Front. Mol. Neurosci. 11 (2018). https://doi.org/10.3389/fnmol.2018.00130.
[80] C.J. Lovitt, T.B. Shelper, V.M. Avery, Doxorubicin resistance in breast cancer cells is mediated by extracellular matrix proteins, BMC Cancer. 18 (2018) 41. https://doi.org/10.1186/s12885-017-3953-6.
[81] J. Lu, X. Ye, F. Fan, L. Xia, R. Bhattacharya, S. Bellister, F. Tozzi, E. Sceusi, Y. Zhou, I. Tachibana, D.M. Maru, D.H. Hawke, J. Rak, S.A. Mani, P. Zweidler-McKay, L.M. Ellis, Endothelial Cells Promote the Colorectal Cancer Stem Cell Phenotype through a Soluble Form of Jagged-1, Cancer Cell. 23 (2013) 171–185. https://doi.org/10.1016/j.ccr.2012.12.021.
[82] J. Landsberg, J. Kohlmeyer, M. Renn, T. Bald, M. Rogava, M. Cron, M. Fatho, V. Lennerz, T. Wölfel, M. Hölzel, T. Tüting, Melanomas resist T-cell therapy through inflammation-induced reversible dedifferentiation, Nature. 490 (2012) 412–416. https://doi.org/10.1038/nature11538.
[83] J. Zhang, J. Liu, Tumor stroma as targets for cancer therapy, Pharmacol. Ther. 137 (2013) 200–215. https://doi.org/10.1016/j.pharmthera.2012.10.003.
149
[84] T.R. Cox, D. Bird, A.M. Baker, H.E. Barker, M.W.Y. Ho, G. Lang, J.T. Erler, LOX-mediated collagen crosslinking is responsible for fibrosis-enhanced metastasis, Cancer Res. 73 (2013) 1721–1732. https://doi.org/10.1158/0008-5472.CAN-12-2233.
[85] A. Winer, S. Adams, P. Mignatti, Matrix metalloproteinase inhibitors in cancer therapy: Turning past failures into future successes, Mol. Cancer Ther. 17 (2018) 1147–1155. https://doi.org/10.1158/1535-7163.MCT-17- 0646.
[86] J. Zhang, S.N. Hochwald, Targeting Receptor Tyrosine Kinases in Solid Tumors, Surg. Oncol. Clin. N. Am. 22 (2013) 685–703. https://doi.org/10.1016/j.soc.2013.06.010.
[87] E. Heller, M.A. Hurchla, J. Xiang, X. Su, S. Chen, J. Schneider, K.S. Joeng, M. Vidal, L. Goldberg, H. Deng, M.C. Hornick, J.L. Prior, D. Piwnica-Worms, F. Long, R. Cagan, K.N. Weilbaecher, Hedgehog signaling inhibition blocks growth of resistant tumors through effects on tumor microenvironment, Cancer Res. 72 (2012) 897–907. https://doi.org/10.1158/0008-5472.CAN-11-2681.
[88] T.A. Gonda, A. Varro, T.C. Wang, B. Tycko, Molecular biology of cancer- associated fibroblasts: Can these cells be targeted in anti-cancer therapy?, Semin. Cell Dev. Biol. 21 (2010) 2–10. https://doi.org/10.1016/j.semcdb.2009.10.001.
[89] R.K. Jain, D.G. Duda, J.W. Clark, J.S. Loeffler, Lessons from phase III clinical trials on anti-VEGF therapy for cancer, Nat. Clin. Pract. Oncol. 3 (2006) 24–40. https://doi.org/10.1038/ncponc0403.
[90] P. Schmid, C. Cruz, F.S. Braiteh, J.P. Eder, S. Tolaney, I. Kuter, R. Nanda, C. Chung, P. Cassier, J.-P. Delord, M. Gordon, Y. Li, B. Liu, C. O’Hear, M. Fasso, L. Molinero, L.A. Emens, Abstract 2986: Atezolizumab in metastatic TNBC (mTNBC): Long-term clinical outcomes and biomarker analyses, in: Cancer Res., American Association for Cancer Research (AACR), 2017: pp. 2986–2986. https://doi.org/10.1158/1538-7445.am2017-2986.
[91] R.R. Rosato, D. Dávila-González, D.S. Choi, W. Qian, W. Chen, A.J. Kozielski, H. Wong, B. Dave, J.C. Chang, Evaluation of anti-PD-1-based therapy against triple-negative breast cancer patient-derived xenograft tumors engrafted in humanized mouse models, Breast Cancer Res. 20 (2018) 108. https://doi.org/10.1186/s13058-018-1037-4.
[92] P. Schmid, S. Adams, H.S. Rugo, A. Schneeweiss, C.H. Barrios, H. Iwata, V. Diéras, R. Hegg, S.-A. Im, G. Shaw Wright, V. Henschel, L. Molinero, S.Y. Chui, R. Funke, A. Husain, E.P. Winer, S. Loi, L.A. Emens, 150
Atezolizumab and Nab-Paclitaxel in Advanced Triple-Negative Breast Cancer, N. Engl. J. Med. 379 (2018) 2108–2121. https://doi.org/10.1056/nejmoa1809615.
[93] E.L.S. Fong, D.A. Harrington, M.C. Farach-Carson, H. Yu, Heralding a new paradigm in 3D tumor modeling, Biomaterials. 108 (2016) 197–213. https://doi.org/10.1016/j.biomaterials.2016.08.052.
[94] E.H. Rubin, D.G. Gilliland, Drug development and clinical trials - The path to an approved cancer drug, Nat. Rev. Clin. Oncol. 9 (2012) 215–222. https://doi.org/10.1038/nrclinonc.2012.22.
[95] K. Louault, T.L. Bonneaud, C. Séveno, P. Gomez-Bougie, F. Nguyen, F. Gautier, N. Bourgeois, D. Loussouarn, O. Kerdraon, S. Barillé-Nion, P. Jézéquel, M. Campone, M. Amiot, P.P. Juin, F. Souazé, Interactions between cancer-associated fibroblasts and tumor cells promote MCL-1 dependency in estrogen receptor-positive breast cancers, Oncogene. 38 (2019) 3261–3273. https://doi.org/10.1038/s41388-018-0635-z.
[96] M. Plaster, S. Singh, H. Tavana, Fibroblasts Promote Proliferation and Matrix Invasion of Breast Cancer Cells in Co‐Culture Models, Adv. Ther. 2 (2019) 1900121. https://doi.org/10.1002/adtp.201900121.
[97] K.E. Sung, X. Su, E. Berthier, C. Pehlke, A. Friedl, D.J. Beebe, Understanding the Impact of 2D and 3D Fibroblast Cultures on In Vitro Breast Cancer Models, PLoS One. 8 (2013) e76373. https://doi.org/10.1371/journal.pone.0076373.
[98] J.P. Gillet, A.M. Calcagno, S. Varma, M. Marino, L.J. Green, M.I. Vora, C. Patel, J.N. Orina, T.A. Eliseeva, V. Singal, R. Padmanabhan, B. Davidson, R. Ganapathi, A.K. Sood, B.R. Rueda, S. V. Ambudkar, M.M. Gottesman, Redefining the relevance of established cancer cell lines to the study of mechanisms of clinical anti-cancer drug resistance, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 18708–18713. https://doi.org/10.1073/pnas.1111840108.
[99] D.W. Hutmacher, D. Loessner, S. Rizzi, D.L. Kaplan, D.J. Mooney, J.A. Clements, Can tissue engineering concepts advance tumor biology research?, Trends Biotechnol. 28 (2010) 125–133. https://doi.org/10.1016/j.tibtech.2009.12.001.
[100] K. Sakamoto, J.W. Schmidt, K.U. Wagner, Mouse models of breast cancer, Methods Mol. Biol. 1267 (2015) 47–71. https://doi.org/10.1007/978-1-4939- 2297-0_3.
[101] L.E. Dobrolecki, S.D. Airhart, D.G. Alferez, S. Aparicio, F. Behbod, M. 151
Bentires-Alj, C. Brisken, C.J. Bult, S. Cai, R.B. Clarke, H. Dowst, M.J. Ellis, E. Gonzalez-Suarez, R.D. Iggo, P. Kabos, S. Li, G.J. Lindeman, E. Marangoni, A. McCoy, F. Meric-Bernstam, H. Piwnica-Worms, M.F. Poupon, J. Reis-Filho, C.A. Sartorius, V. Scabia, G. Sflomos, Y. Tu, F. Vaillant, J.E. Visvader, A. Welm, M.S. Wicha, M.T. Lewis, Patient-derived xenograft (PDX) models in basic and translational breast cancer research, Cancer Metastasis Rev. 35 (2016) 547–573. https://doi.org/10.1007/s10555-016-9653-x.
[102] H.S. Park, J.D. Lee, J.Y. Kim, S. Park, J.H. Kim, H.J. Han, Y.A. Choi, A.R. Choi, J.H. Sohn, S. Il Kim, Establishment of chemosensitivity tests in triple- negative and BRCA-mutated breast cancer patient-derived xenograft models, PLoS One. 14 (2019). https://doi.org/10.1371/journal.pone.0225082.
[103] E. Iorns, K. Drews-Elger, T.M. Ward, S. Dean, J. Clarke, D. Berry, D. El Ashry, M. Lippman, A New Mouse Model for the Study of Human Breast Cancer Metastasis, PLoS One. 7 (2012) e47995. https://doi.org/10.1371/journal.pone.0047995.
[104] I. Primac, E. Maquoi, S. Blacher, R. Heljasvaara, J. van Deun, H.Y.H. Smeland, A. Canale, T. Louis, L. Stuhr, N.E. Sounni, D. Cataldo, T. Pihlajaniemi, C. Pequeux, O. de Wever, D. Gullberg, A. Noel, Stromal integrin α11 regulates PDGFRβ signaling and promotes breast cancer progression, J. Clin. Invest. 129 (2019) 4609–4628. https://doi.org/10.1172/JCI125890.
[105] K. Takai, A. Le, V.M. Weaver, Z. Werb, Targeting the cancer-associated fibroblasts as a treatment in triple-negative breast cancer, Oncotarget. 7 (2016) 82889–82901. https://doi.org/10.18632/oncotarget.12658.
[106] M. Huang, Y. Li, H. Zhang, F. Nan, Breast cancer stromal fibroblasts promote the generation of CD44 +CD24- cells through SDF-1/CXCR4 interaction, J. Exp. Clin. Cancer Res. 29 (2010). https://doi.org/10.1186/1756-9966-29-80.
[107] C.R. Ireson, M.S. Alavijeh, A.M. Palmer, E.R. Fowler, H.J. Jones, The role of mouse tumour models in the discovery and development of anticancer drugs, Br. J. Cancer. 121 (2019) 101–108. https://doi.org/10.1038/s41416- 019-0495-5.
[108] G.S. Herter-Sprie, A.L. Kung, K.K. Wong, New cast for a new era: Preclinical cancer drug development revisited, J. Clin. Invest. 123 (2013) 3636–3645. https://doi.org/10.1172/JCI68340.
[109] I. Holen, V. Speirs, B. Morrissey, K. Blyth, In vivo models in breast cancer 152
research: Progress, challenges and future directions, DMM Dis. Model. Mech. 10 (2017) 359–371. https://doi.org/10.1242/dmm.028274.
[110] N. Peela, D. Truong, H. Saini, H. Chu, S. Mashaghi, S.L. Ham, S. Singh, H. Tavana, B. Mosadegh, M. Nikkhah, Advanced biomaterials and microengineering technologies to recapitulate the stepwise process of cancer metastasis, Biomaterials. 133 (2017) 176–207. https://doi.org/10.1016/j.biomaterials.2017.04.017.
[111] D.W. Infanger, M.E. Lynch, C. Fischbach, Engineered culture models for studies of tumor-microenvironment interactions, Annu. Rev. Biomed. Eng. 15 (2013) 29–53. https://doi.org/10.1146/annurev-bioeng-071811-150028.
[112] S. Singh, S. Tran, J. Putman, H. Tavana, Three-dimensional models of breast cancer–fibroblasts interactions, Exp. Biol. Med. 245 (2020) 879– 888. https://doi.org/10.1177/1535370220917366.
[113] S. Nath, G.R. Devi, Three-dimensional culture systems in cancer research: Focus on tumor spheroid model, Pharmacol. Ther. 163 (2016) 94–108. https://doi.org/10.1016/j.pharmthera.2016.03.013.
[114] V. Brancato, J.M. Oliveira, V.M. Correlo, R.L. Reis, S.C. Kundu, Could 3D models of cancer enhance drug screening?, Biomaterials. 232 (2020) 119744. https://doi.org/10.1016/j.biomaterials.2019.119744.
[115] P.S. Thakuri, G.D. Luker, H. Tavana, Cyclical Treatment of Colorectal Tumor Spheroids Induces Resistance to MEK Inhibitors, Transl. Oncol. 12 (2019) 404–416. https://doi.org/10.1016/j.tranon.2018.11.009.
[116] S. Modok, H.R. Mellor, R. Callaghan, Modulation of multidrug resistance efflux pump activity to overcome chemoresistance in cancer, Curr. Opin. Pharmacol. 6 (2006) 350–354. https://doi.org/10.1016/j.coph.2006.01.009.
[117] M. Zanoni, F. Piccinini, C. Arienti, A. Zamagni, S. Santi, R. Polico, A. Bevilacqua, A. Tesei, 3D tumor spheroid models for in vitro therapeutic screening: A systematic approach to enhance the biological relevance of data obtained, Sci. Rep. 6 (2016) 19103. https://doi.org/10.1038/srep19103.
[118] H. Shao, M. Moller, D. Wang, A. Ting, M. Boulina, Z.J. Liu, A novel stromal fibroblast-modulated 3d tumor spheroid model for studying tumor-stroma interaction and drug discovery, J. Vis. Exp. 2020 (2020). https://doi.org/10.3791/60660.
[119] E. Fennema, N. Rivron, J. Rouwkema, C. van Blitterswijk, J. De Boer, Spheroid culture as a tool for creating 3D complex tissues, Trends 153
Biotechnol. 31 (2013) 108–115. https://doi.org/10.1016/j.tibtech.2012.12.003.
[120] X. Cui, Y. Hartanto, H. Zhang, Advances in multicellular spheroids formation, J. R. Soc. Interface. 14 (2017). https://doi.org/10.1098/rsif.2016.0877.
[121] G. Lazzari, V. Nicolas, M. Matsusaki, M. Akashi, P. Couvreur, S. Mura, Multicellular spheroid based on a triple co-culture: A novel 3D model to mimic pancreatic tumor complexity, Acta Biomater. 78 (2018) 296–307. https://doi.org/10.1016/j.actbio.2018.08.008.
[122] S.A. Kim, E.K. Lee, H.J. Kuh, Co-culture of 3D tumor spheroids with fibroblasts as a model for epithelial-mesenchymal transition in vitro, Exp. Cell Res. 335 (2015) 187–196. https://doi.org/10.1016/j.yexcr.2015.05.016.
[123] M. Majety, L.P. Pradel, M. Gies, C.H. Ries, Fibroblasts influence survival and therapeutic response in a 3D co-culture model, PLoS One. 10 (2015). https://doi.org/10.1371/journal.pone.0127948.
[124] L.A. Kunz-Schughart, M. Kreutz, R. Knuechel, Multicellular spheroids: A three-dimensional in vitro culture system to study tumour biology, Int. J. Exp. Pathol. 79 (1998) 1–23. https://doi.org/10.1046/j.1365- 2613.1998.00051.x.
[125] R.Z. Lin, H.Y. Chang, Recent advances in three-dimensional multicellular spheroid culture for biomedical research, Biotechnol. J. 3 (2008) 1172– 1184. https://doi.org/10.1002/biot.200700228.
[126] B.R. Whatley, X. Li, N. Zhang, X. Wen, Magnetic-directed patterning of cell spheroids, J. Biomed. Mater. Res. - Part A. 102 (2014) 1537–1547. https://doi.org/10.1002/jbm.a.34797.
[127] M. Vinci, S. Gowan, F. Boxall, L. Patterson, M. Zimmermann, W. Court, C. Lomas, M. Mendiola, D. Hardisson, S.A. Eccles, Advances in establishment and analysis of three-dimensional tumor spheroid-based functional assays for target validation and drug evaluation, BMC Biol. 10 (2012). https://doi.org/10.1186/1741-7007-10-29.
[128] E. Atefi, S. Lemmo, D. Fyffe, G.D. Luker, H. Tavana, High throughput, polymeric aqueous two-phase printing of tumor spheroids, Adv. Funct. Mater. 24 (2014) 6509–6515. https://doi.org/10.1002/adfm.201401302.
[129] S. Lemmo, E. Atefi, G.D. Luker, H. Tavana, Optimization of aqueous biphasic tumor spheroid microtechnology for anti-cancer drug testing in 3D culture, Cell. Mol. Bioeng. 7 (2014) 344–354. 154
https://doi.org/10.1007/s12195-014-0349-4.
[130] J.E. Ekert, K. Johnson, B. Strake, J. Pardinas, S. Jarantow, R. Perkinson, D.C. Colter, Three-dimensional lung tumor microenvironment modulates therapeutic compound responsiveness in vitro - Implication for drug development, PLoS One. 9 (2014). https://doi.org/10.1371/journal.pone.0092248.
[131] S. Sant, P.A. Johnston, The production of 3D tumor spheroids for cancer drug discovery, Drug Discov. Today Technol. 23 (2017) 27–36. https://doi.org/10.1016/j.ddtec.2017.03.002.
[132] P.S. Thakuri, M. Gupta, M. Plaster, H. Tavana, Quantitative Size-Based Analysis of Tumor Spheroids and Responses to Therapeutics, Assay Drug Dev. Technol. 17 (2019) 140–149. https://doi.org/10.1089/adt.2018.895.
[133] S.L. Ham, P.S. Thakuri, M. Plaster, J. Li, K.E. Luker, G.D. Luker, H. Tavana, Three-dimensional tumor model mimics stromal-breast cancer cells signaling, Oncotarget. 9 (2018) 249. https://doi.org/10.18632/oncotarget.22922.
[134] G. Johansson, Partition of salts and their effects on partition of proteins in a dextran-poly(ethylene glycol)-water two-phase system, BBA - Protein Struct. 221 (1970) 387–390. https://doi.org/10.1016/0005-2795(70)90283- 7.
[135] J.R. SooHoo, G.M. Walker, Microfluidic aqueous two phase system for leukocyte concentration from whole blood, Biomed. Microdevices. 11 (2009) 323–329. https://doi.org/10.1007/s10544-008-9238-8.
[136] H. Tavana, A. Jovic, B. Mosadegh, Q.Y. Lee, X. Liu, K.E. Luker, G.D. Luker, S.J. Weiss, S. Takayama, Nanolitre liquid patterning in aqueous environments for spatially defined reagent delivery to mammalian cells, Nat. Mater. 8 (2009) 736–741. https://doi.org/10.1038/nmat2515.
[137] A.S. Schmidt, A.M. Ventom, J.A. Asenjo, Partitioning and purification of α- amylase in aqueous two-phase systems, Enzyme Microb. Technol. 16 (1994) 131–142. https://doi.org/10.1016/0141-0229(94)90076-0.
[138] M. Iqbal, Y. Tao, S. Xie, Y. Zhu, D. Chen, X. Wang, L. Huang, D. Peng, A. Sattar, M.A.B. Shabbir, H.I. Hussain, S. Ahmed, Z. Yuan, Aqueous two- phase system (ATPS): an overview and advances in its applications, Biol. Proced. Online. 18 (2016) 1–18. https://doi.org/10.1186/s12575-016-0048- 8.
[139] E. Atefi, R. Joshi, J.A. Mann, H. Tavana, Interfacial Tension Effect on Cell 155
Partition in Aqueous Two-Phase Systems., ACS Appl. Mater. Interfaces. 7 (2015) 21305–21314. https://doi.org/10.1021/acsami.5b05757.
[140] N. Peela, F.S. Sam, W. Christenson, D. Truong, A.W. Watson, G. Mouneimne, R. Ros, M. Nikkhah, A three dimensional micropatterned tumor model for breast cancer cell migration studies, Biomaterials. 81 (2016) 72–83. https://doi.org/10.1016/j.biomaterials.2015.11.039.
[141] G.P. Raeber, M.P. Lutolf, J.A. Hubbell, Molecularly engineered PEG hydrogels: A novel model system for proteolytically mediated cell migration, Biophys. J. 89 (2005) 1374–1388. https://doi.org/10.1529/biophysj.104.050682.
[142] F. Bersani, J. Lee, M. Yu, R. Morris, R. Desai, S. Ramaswamy, M. Toner, D.A. Haber, B. Parekkadan, Bioengineered implantable scaffolds as a tool to study stromal-derived factors in metastatic cancer models, Cancer Res. 74 (2014) 7229–7238. https://doi.org/10.1158/0008-5472.CAN-14-1809.
[143] V. Phan-Lai, S.J. Florczyk, F.M. Kievit, K. Wang, E. Gad, M.L. Disis, M. Zhang, Three-dimensional scaffolds to evaluate tumor associated fibroblast-mediated suppression of breast tumor specific T cells, Biomacromolecules. 14 (2013) 1330–1337. https://doi.org/10.1021/bm301928u.
[144] S.K. Sahoo, A.K. Panda, V. Labhasetwar, Characterization of porous PLGA/PLA microparticles as a scaffold for three dimensional growth of breast cancer cells, Biomacromolecules. 6 (2005) 1132–1139. https://doi.org/10.1021/bm0492632.
[145] Z.M. Che, T.H. Jung, J.H. Choi, D.J. Yoon, H.J. Jeong, E.J. Lee, J. Kim, Collagen-based co-culture for invasive study on cancer cells-fibroblasts interaction, Biochem. Biophys. Res. Commun. 346 (2006) 268–275. https://doi.org/10.1016/j.bbrc.2006.05.111.
[146] Q. Cui, B. Wang, K. Li, H. Sun, T. Hai, Y. Zhang, H. Kang, Upregulating MMP‐1 in carcinoma‐associated fibroblasts reduces the efficacy of taxotere on breast cancer synergized by collagen IV, Oncol. Lett. 16 (2018) 3537– 3544. https://doi.org/10.3892/ol.2018.9092.
[147] C.S. Hughes, L.M. Postovit, G.A. Lajoie, Matrigel: a complex protein mixture required for optimal growth of cell culture., Proteomics. 10 (2010) 1886–1890. https://doi.org/10.1002/pmic.200900758.
[148] J. Liu, P. Li, L. Wang, M. Li, Z. Ge, L. Noordam, R. Lieshout, M.M.A. Verstegen, B. Ma, J. Su, Q. Yang, R. Zhang, G. Zhou, L.C. Carrascosa, D. Sprengers, J.N.M. IJzermans, R. Smits, J. Kwekkeboom, L.J.W. van der 156
Laan, M.P. Peppelenbosch, Q. Pan, W. Cao, Cancer-Associated Fibroblasts Provide a Stromal Niche for Liver Cancer Organoids That Confers Trophic Effects and Therapy Resistance, Cell. Mol. Gastroenterol. Hepatol. (2020). https://doi.org/10.1016/j.jcmgh.2020.09.003.
[149] D. Öhlund, A. Handly-Santana, G. Biffi, E. Elyada, A.S. Almeida, M. Ponz- Sarvise, V. Corbo, T.E. Oni, S.A. Hearn, E.J. Lee, I.I.C. Chio, C. Il Hwang, H. Tiriac, L.A. Baker, D.D. Engle, C. Feig, A. Kultti, M. Egeblad, D.T. Fearon, J.M. Crawford, H. Clevers, Y. Park, D.A. Tuveson, Distinct populations of inflammatory fibroblasts and myofibroblasts in pancreatic cancer, J. Exp. Med. 214 (2017) 579–596. https://doi.org/10.1084/jem.20162024.
[150] K. Stock, M.F. Estrada, S. Vidic, K. Gjerde, A. Rudisch, V.E. Santo, M. Barbier, S. Blom, S.C. Arundkar, I. Selvam, A. Osswald, Y. Stein, S. Gruenewald, C. Brito, W. Van Weerden, V. Rotter, E. Boghaert, M. Oren, W. Sommergruber, Y. Chong, R. De Hoogt, R. Graeser, Capturing tumor complexity in vitro: Comparative analysis of 2D and 3D tumor models for drug discovery, Sci. Rep. 6 (2016) 1–15. https://doi.org/10.1038/srep28951.
[151] Y. Fang, R.M. Eglen, Three-Dimensional Cell Cultures in Drug Discovery and Development, SLAS Discov. 22 (2017) 456–472. https://doi.org/10.1177/1087057117696795.
[152] I. Villanueva, C.A. Weigel, S.J. Bryant, Cell-matrix interactions and dynamic mechanical loading influence chondrocyte gene expression and bioactivity in PEG-RGD hydrogels, Acta Biomater. 5 (2009) 2832–2846. https://doi.org/10.1016/j.actbio.2009.05.039.
[153] X. Yue, T.D. Nguyen, V. Zellmer, S. Zhang, P. Zorlutuna, Stromal cell- laden 3D hydrogel microwell arrays as tumor microenvironment model for studying stiffness dependent stromal cell-cancer interactions, Biomaterials. 170 (2018) 37–48. https://doi.org/10.1016/j.biomaterials.2018.04.001.
[154] Z. Liu, G. Vunjak-Novakovic, Modeling tumor microenvironments using custom-designed biomaterial scaffolds, Curr. Opin. Chem. Eng. 11 (2016) 94–105. https://doi.org/10.1016/j.coche.2016.01.012.
[155] G. Rijal, W. Li, 3D scaffolds in breast cancer research, Biomaterials. 81 (2016) 135–156. https://doi.org/10.1016/j.biomaterials.2015.12.016.
[156] K. Hockemeyer, C. Janetopoulos, A. Terekhov, W. Hofmeister, A. Vilgelm, L. Costa, J.P. Wikswo, A. Richmond, Engineered three-dimensional microfluidic device for interrogating cell-cell interactions in the tumor microenvironment, Biomicrofluidics. 8 (2014) 044105. 157
https://doi.org/10.1063/1.4890330.
[157] D.D. Truong, A. Kratz, J.G. Park, E.S. Barrientos, H. Saini, T. Nguyen, B. Pockaj, G. Mouneimne, J. LaBaer, M. Nikkhah, A human organotypic microfluidic tumor model permits investigation of the interplay between patient-derived fibroblasts and breast cancer cells, Cancer Res. 79 (2019) 3139–3151. https://doi.org/10.1158/0008-5472.can-18-2293.
[158] K.E. Sung, N. Yang, C. Pehlke, P.J. Keely, K.W. Eliceiri, A. Friedl, D.J. Beebe, Transition to invasion in breast cancer: A microfluidic in vitro model enables examination of spatial and temporal effects, Integr. Biol. 3 (2011) 439–450. https://doi.org/10.1039/c0ib00063a.
[159] F. Gioiella, F. Urciuolo, G. Imparato, V. Brancato, P.A. Netti, An Engineered Breast Cancer Model on a Chip to Replicate ECM-Activation In Vitro during Tumor Progression, Adv. Healthc. Mater. 5 (2016) 3074–3084. https://doi.org/10.1002/adhm.201600772.
[160] J.W. Song, S.P. Cavnar, A.C. Walker, K.E. Luker, M. Gupta, Y.C. Tung, G.D. Luker, S. Takayama, Microfluidic endothelium for studying the intravascular adhesion of metastatic breast cancer cells, PLoS One. 4 (2009) e5756. https://doi.org/10.1371/journal.pone.0005756.
[161] S. Halldorsson, E. Lucumi, R. Gómez-Sjöberg, R.M.T. Fleming, Advantages and challenges of microfluidic cell culture in polydimethylsiloxane devices, Biosens. Bioelectron. 63 (2015) 218–231. https://doi.org/10.1016/j.bios.2014.07.029.
[162] C.W. Chi, Y.H. Lao, A.H.R. Ahmed, E.C. Benoy, C. Li, Z. Dereli-Korkut, B.M. Fu, K.W. Leong, S. Wang, High-Throughput Tumor-on-a-Chip Platform to Study Tumor–Stroma Interactions and Drug Pharmacokinetics, Adv. Healthc. Mater. 9 (2020). https://doi.org/10.1002/adhm.202000880.
[163] G. Du, Q. Fang, J.M.J. den Toonder, Microfluidics for cell-based high throughput screening platforms-A review, Anal. Chim. Acta. 903 (2016) 36– 50. https://doi.org/10.1016/j.aca.2015.11.023.
[164] E. Atefi, J.A. Mann, H. Tavana, Ultralow interfacial tensions of aqueous two-phase systems measured using drop shape, Langmuir. 30 (2014) 9691–9699. https://doi.org/10.1021/la500930x.
[165] J.A. Hickman, R. Graeser, R. de Hoogt, S. Vidic, C. Brito, M. Gutekunst, H. van der Kuip, Imi Predect consortium, Three-dimensional models of cancer for pharmacology and cancer cell biology: Capturing tumor complexity in vitro/ex vivo, Biotechnol. J. 9 (2014) 1115–1128. https://doi.org/10.1002/biot.201300492. 158
[166] M.R. Carvalho, D. Lima, R.L. Reis, V.M. Correlo, J.M. Oliveira, Evaluating Biomaterial- and Microfluidic-Based 3D Tumor Models, Trends Biotechnol. 33 (2015) 667–678. https://doi.org/10.1016/j.tibtech.2015.09.009.
[167] P. Saglam-Metiner, S. Gulce-Iz, C. Biray-Avci, Bioengineering-inspired three-dimensional culture systems: Organoids to create tumor microenvironment, Gene. 686 (2019) 203–212. https://doi.org/10.1016/j.gene.2018.11.058.
[168] L.M. Granat, O. Kambhampati, S. Klosek, B. Niedzwecki, K. Parsa, D. Zhang, The promises and challenges of patient‐derived tumor organoids in drug development and precision oncology, Anim. Model. Exp. Med. 2 (2019) 150–161. https://doi.org/10.1002/ame2.12077.
[169] P.Å. Albertsson, Fractionation of particles and macromolecules in aqueous two-phase systems, Biochem. Pharmacol. 5 (1961) 351–358. https://doi.org/10.1016/0006-2952(61)90028-4.
[170] H. Walter, C. Larsson, Partitioning procedures and techniques: Cells, organelles, and membranes, Methods Enzymol. 228 (1994) 42–63. https://doi.org/10.1016/0076-6879(94)28006-1.
[171] M. Li, J.W. Kim, T.L. Peeples, Amylase partitioning and extractive bioconversion of starch using thermoseparating aqueous two-phase systems, J. Biotechnol. 93 (2002) 15–26. https://doi.org/10.1016/S0168- 1656(01)00382-0.
[172] M.R. Gavasane, V.G. Gaikar, Aqueous two-phase affinity partitioning of penicillin acylase from E. Coli in presence of PEG-derivatives, Enzyme Microb. Technol. 32 (2003) 665–675. https://doi.org/10.1016/S0141- 0229(03)00032-2.
[173] C. Moraes, A.B. Simon, A.J. Putnam, S. Takayama, Aqueous two-phase printing of cell-containing contractile collagen microgels, Biomaterials. 34 (2013) 9623–9631. https://doi.org/10.1016/j.biomaterials.2013.08.046.
[174] P.A. Albertsson, F. Tjerneld, Phase diagrams, Methods Enzymol. 228 (1994) 3–13. https://doi.org/10.1016/0076-6879(94)28003-7.
[175] R.E. NEUMAN, M.A. LOGAN, The determination of hydroxyproline., J. Biol. Chem. 184 (1950) 299–306. https://doi.org/10.1016/0009-8981(65)90038- 0.
[176] E. Atefi, D. Fyffe, K.B. Kaylan, H. Tavana, Characterization of Aqueous Two-Phase Systems from Volume and Density Measurements, J. Chem. Eng. Data. 61 (2016) 1531–1539. 159
https://doi.org/10.1021/acs.jced.5b00901.
[177] S.L. Ham, R. Joshi, G.D. Luker, H. Tavana, Engineered Breast Cancer Cell Spheroids Reproduce Biologic Properties of Solid Tumors, Adv. Healthc. Mater. 5 (2016) 2788–2798. https://doi.org/10.1002/adhm.201600644.
[178] P.-P. Albertsson, A. Cajarville, D.E. Brooks, F. Tjerneld, Partition of proteins in aqueous polymer two-phase systems and the effect of molecular weight of the polymer, Biochim. Biophys. Acta. 926 (1987) 87– 93. https://doi.org/http://dx.doi.org/10.1016/0304-4165(87)90185-1.
[179] Y. Liang, J. Jeong, R.J. DeVolder, C. Cha, F. Wang, Y.W. Tong, H. Kong, A cell-instructive hydrogel to regulate malignancy of 3D tumor spheroids with matrix rigidity, Biomaterials. 32 (2011) 9308–9315. https://doi.org/10.1016/j.biomaterials.2011.08.045.
[180] D. Yip, C.H. Cho, A multicellular 3D heterospheroid model of liver tumor and stromal cells in collagen gel for anti-cancer drug testing, Biochem. Biophys. Res. Commun. 433 (2013) 327–332. https://doi.org/10.1016/j.bbrc.2013.03.008.
[181] P. Albertsson, Partition of cell particles and macromolecules in polymer two-phase systems, Adv. Protein Chem. 24 (1970) 309–341. https://doi.org/10.1016/S0065-3233(08)60244-2.
[182] M. Plodinec, M. Loparic, C.A. Monnier, E.C. Obermann, R. Zanetti- Dallenbach, P. Oertle, J.T. Hyotyla, U. Aebi, M. Bentires-Alj, R.Y.H. Lim, C.A. Schoenenberger, The nanomechanical signature of breast cancer, Nat. Nanotechnol. 7 (2012) 757–765. https://doi.org/10.1038/nnano.2012.167.
[183] M. Miron-Mendoza, J. Seemann, F. Grinnell, The differential regulation of cell motile activity through matrix stiffness and porosity in three dimensional collagen matrices, Biomaterials. 31 (2010) 6425–6435. https://doi.org/10.1016/j.biomaterials.2010.04.064.
[184] S. Ramanujan, A. Pluen, T.D. McKee, E.B. Brown, Y. Boucher, R.K. Jain, Diffusion and convection in collagen gels: Implications for transport in the tumor interstitium, Biophys. J. 83 (2002) 1650–1660. https://doi.org/10.1016/S0006-3495(02)73933-7.
[185] P.S. Thakuri, S.L. Ham, G.D. Luker, H. Tavana, Multiparametric Analysis of Oncology Drug Screening with Aqueous Two-Phase Tumor Spheroids, Mol. Pharm. 13 (2016) 3724–3735. https://doi.org/10.1021/acs.molpharmaceut.6b00527.
160
[186] R. Kalluri, M. Zeisberg, Fibroblasts in cancer, Nat. Rev. Cancer. 6 (2006) 392–401. https://doi.org/10.1038/nrc1877.
[187] R. Kalluri, The biology and function of exosomes in cancer, J. Clin. Invest. (2016) 1208–1215. https://doi.org/10.1172/JCI81135.
[188] E.M. De Kruijf, J.G.H. Van Nes, C.J.H. Van De Velde, H. Putter, V.T.H.B.M. Smit, G.J. Liefers, P.J.K. Kuppen, R.A.E.M. Tollenaar, W.E. Mesker, Tumor-stroma ratio in the primary tumor is a prognostic factor in early breast cancer patients, especially in triple-negative carcinoma patients, Breast Cancer Res. Treat. 125 (2011) 687–696. https://doi.org/10.1007/s10549-010-0855-6.
[189] T. Marsh, K. Pietras, S. S. McAllister, Fibroblasts as architects of cancer pathogenesis, Biochim. Biophys. Acta - Mol. Basis Dis. 1832 (2013) 1070– 1078. https://doi.org/10.1016/j.bbadis2012.10.013.
[190] L. Ronnov-Jessen, A function for filamentous alpha-smooth muscle actin: retardation of motility in fibroblasts, J. Cell Biol. 134 (2004) 67–80. https://doi.org/10.1083/jcb.134.1.67.
[191] Y. Attieh, D.M. Vignjevic, The hallmarks of CAFs in cancer invasion, Eur. J. Cell Biol. 95 (2016) 493–502. https://doi.org/10.1016/j.ejcb.2016.07.004.
[192] H. Ungefroren, S. Sebens, D. Seidl, H. Lehnert, R. Hass, Interaction of tumor cells with the microenvironment, Cell Commun. Signal. 9 (2011) 18. https://doi.org/10.1186/1478-811X-9-18.
[193] G.Y. Lee, P.A. Kenny, E.H. Lee, M.J. Bissell, Three-dimensional culture models of normal and malignant breast epithelial cells, Nat. Methods. 4 (2007) 359–365. https://doi.org/10.1038/nmeth1015.
[194] B. Weigelt, A.T. Lo, C.C. Park, J.W. Gray, M.J. Bissell, HER2 signaling pathway activation and response of breast cancer cells to HER2-targeting agents is dependent strongly on the 3D microenvironment, Breast Cancer Res. Treat. 122 (2010) 35–43. https://doi.org/10.1007/s10549-009-0502-2.
[195] J.R. Bradford, M. Wappett, G. Beran, A. Logie, O. Delpuech, H. Brown, J. Boros, N.J. Camp, R. McEwen, A.M. Mazzola, C. D’Cruz, S.T. Barry, Whole transcriptome profiling of patient-derived xenograft models as a tool to identify both tumor and stromal specific biomarkers, Oncotarget. 7 (2016) 20773. https://doi.org/10.18632/oncotarget.8014.
[196] G.H. Lee, J.S. Lee, X. Wang, S.H. Lee, Bottom-Up Engineering of Well- Defined 3D Microtissues Using Microplatforms and Biomedical Applications, Adv. Healthc. Mater. 5 (2016) 56–74. 161
https://doi.org/10.1002/adhm.201500107.
[197] I.M. Andrew, F.T. Ian, Drug penetration in solid tumours, Nat. Rev. Cancer. 6 (2006) 583–592. https://doi.org/10.1038/nrc1893.
[198] H. Yamaguchi, N. Yoshida, M. Takanashi, Y. Ito, K. Fukami, K. Yanagihara, M. Yashiro, R. Sakai, Stromal fibroblasts mediate extracellular matrix remodeling and invasion of scirrhous gastric carcinoma cells, PLoS One. 9 (2014) e85485. https://doi.org/10.1371/journal.pone.0085485.
[199] D. Izumi, T. Ishimoto, K. Miyake, H. Sugihara, K. Eto, H. Sawayama, T. Yasuda, Y. Kiyozumi, T. Kaida, J. Kurashige, Y. Imamura, Y. Hiyoshi, M. Iwatsuki, S. Iwagami, Y. Baba, Y. Sakamoto, Y. Miyamoto, N. Yoshida, M. Watanabe, H. Takamori, N. Araki, P. Tan, H. Baba, CXCL12/CXCR4 activation by cancer-associated fibroblasts promotes integrin β1 clustering and invasiveness in gastric cancer, Int. J. Cancer. 138 (2016) 1207–1219. https://doi.org/10.1002/ijc.29864.
[200] J.H. Lee, S.K. Kim, I.A. Khawar, S.Y. Jeong, S. Chung, H.J. Kuh, Microfluidic co-culture of pancreatic tumor spheroids with stellate cells as a novel 3D model for investigation of stroma-mediated cell motility and drug resistance, J. Exp. Clin. Cancer Res. 37 (2018) 4. https://doi.org/10.1186/s13046-017-0654-6.
[201] Y. Xu, S. Mi, Z. Wu, X. Qian, Z. Du, M. Zhang, W. Sun, Microfluidic co- culture system for cancer migratory analysis and anti-metastatic drugs screening, Sci. Rep. 6 (2016) 1–11. https://doi.org/10.1038/srep35544.
[202] N. V. Menon, Y.J. Chuah, B. Cao, M. Lim, Y. Kang, A microfluidic co- culture system to monitor tumor-stromal interactions on a chip, Biomicrofluidics. 8 (2014) 064118. https://doi.org/10.1063/1.4903762.
[203] J.D. Wang, N.J. Douville, S. Takayama, M. Elsayed, Quantitative analysis of molecular absorption into PDMS microfluidic channels, Ann. Biomed. Eng. 40 (2012) 1862–1873. https://doi.org/10.1007/s10439-012-0562-z.
[204] J.C. McDonald, D.C. Duffy, J.R. Anderson, D.T. Chiu, H. Wu, O.J.A. Schueller, G.M. Whitesides, Fabrication of microfluidic systems in poly(dimethylsiloxane), An Int. J. 21 (2000) 27–40. https://doi.org/10.1002/(SICI)1522-2683(20000101)21:1<27::AID- ELPS27>3.0.CO;2-C.
[205] K.E. Luker, G.D. Luker, Functions of CXCL12 and CXCR4 in breast cancer, Cancer Lett. 238 (2006) 30–41. https://doi.org/10.1016/j.canlet.2005.06.021.
162
[206] M.C.P. Smith, K.E. Luker, J.R. Garbow, J.L. Prior, E. Jackson, D. Piwnica- Worms, G.D. Luker, CXCR4 regulates growth of both primary and metastatic breast cancer, Cancer Res. 64 (2004) 8604–8612. https://doi.org/10.1158/0008-5472.CAN-04-1844.
[207] S. Singh, H. Tavana, Collagen Partition in Polymeric Aqueous Two-Phase Systems for Tissue Engineering, Front. Chem. 6 (2018). https://doi.org/10.3389/fchem.2018.00379.
[208] S.L. Ham, E. Atefi, D. Fyffe, H. Tavana, Robotic production of cancer cell spheroids with an aqueous two-phase system for drug testing, J. Vis. Exp. 2015 (2015) e52754. https://doi.org/10.3791/52754.
[209] L. Ronnov-Jessen, O.W. Petersen, M.J. Bissell, Cellular changes involved in conversion of normal to malignant breast: importance of the stromal reaction, Physiol. Rev. 76 (2017) 69–125. https://doi.org/10.1152/physrev.1996.76.1.69.
[210] J. Wu, C. Liang, M. Chen, W. Su, Association between tumor-stroma ratio and prognosis in solid tumor patients: a systematic review and meta- analysis, Oncotarget. 7 (2016). https://doi.org/10.18632/oncotarget.12135.
[211] C.J.H. Kramer, K.M.H. Vangangelt, G.W. van Pelt, T.J.A. Dekker, R.A.E.M. Tollenaar, W.E. Mesker, The prognostic value of tumour–stroma ratio in primary breast cancer with special attention to triple-negative tumours: a review, Breast Cancer Res. Treat. 173 (2019) 55–64. https://doi.org/10.1007/s10549-018-4987-4.
[212] F. Grinnell, Fibroblast biology in 3D collagen matrices, Trends Cell Biol. 13 (2003) 264–269. https://doi.org/10.1016/S0962-8924(03)00057-6.
[213] E. Bell, B. Ivarsson, C. Merrill, Production of a tissue-like structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vitro., Proc. Natl. Acad. Sci. U. S. A. 76 (1979) 1274–8. https://doi.org/10.1073/pnas.76.3.1274.
[214] Y.K. Zhu, T. Umino, X.D. Liu, H.J. Wang, D.J. Romberger, J.R. Spurzem, S.I. Rennard, Contraction of fibroblast-containing collagen gels: Initial collagen concentration regulates the degree of contraction and cell survival, Vitr. Cell. Dev. Biol. - Anim. 37 (2002) 10. https://doi.org/10.1290/1071-2690(2001)037<0010:cofccg>2.0.co;2.
[215] J.J. Tomasek, C.J. Haaksma, R.J. Eddy, M.B. Vaughan, Fibroblast contraction occurs on release of tension in attached collagen lattices: Dependency on an organized actin cytoskeleton and serum, Anat. Rec. 232 (1992) 359–368. https://doi.org/10.1002/ar.1092320305. 163
[216] K. Burridge, Are stress fibres contractile?, Nature. 294 (1981) 691–692. https://doi.org/10.1038/294691a0.
[217] K. Katoh, Y. Kano, M. Masuda, H. Onishi, K. Fujiwara, Isolation and Contraction of the Stress Fiber, Mol. Biol. Cell. 9 (1998) 1919–1938. https://doi.org/10.1091/mbc.9.7.1919.
[218] A. Yuge, K. Nasu, H. Matsumoto, M. Nishida, H. Narahara, Collagen gel contractility is enhanced in human endometriotic stromal cells: A possible mechanism underlying the pathogenesis of endometriosis- associated fibrosis, Hum. Reprod. 22 (2007) 938–944. https://doi.org/10.1093/humrep/del485.
[219] J.J. Tomasek, M.B. Vaughan, B.P. Kropp, G. Gabbiani, M.D. Martin, C.J. Haaksma, B. Hinz, Contraction of myofibroblasts in granulation tissue is dependent on Rho/Rho kinase/myosin light chain phosphatase activity, Wound Repair Regen. 14 (2006) 313–320. https://doi.org/10.1111/j.1743- 6109.2006.00126.x.
[220] M. Kolodney, E.. Elson, Correlation of myosin light chain phosphorylation with isometric contraction of fibroblasts., J. Biol. Chem. 32 (1993) 23850– 23855.
[221] S. Hooper, C. Hidalgo-Carcedo, E. Sahai, C. Gaggioli, K. Harrington, R. Grosse, J.F. Marshall, Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells, Nat. Cell Biol. 9 (2007) 1392–1400. https://doi.org/10.1038/ncb1658.
[222] M.F. Olson, E. Sahai, The actin cytoskeleton in cancer cell motility, 2009. https://doi.org/10.1007/s10585-008-9174-2.
[223] O. De Wever, Q.D. Nguyen, L. Van Hoorde, M. Bracke, E. Bruyneel, C. Gespach, M. Mareel, Tenascin-C and SF/HGF produced by myofibroblasts in vitro provide convergent pro-invasive signals to human colon cancer cells through RhoA and Rac, FASEB J. 18 (2004) 1016–1018. https://doi.org/10.1096/fj.03-1110fje.
[224] S. Fiering, L.-H. Ang, J. Lacoste, T.D. Smith, E. Griner, Registered report: Biomechanical remodeling of the microenvironment by stromal caveolin-1 favors tumor invasion and metastasis, Elife. 4 (2015). https://doi.org/10.7554/elife.04796.
[225] J.. Goetz, S. Minguet, I. Navarro-Lérida, J.. Lazcano, R. Samaniego, E. Calvo, M. Tello, T. Osteso-Ibáñez, T. Pellinen, A. Echarri, A. Cerezo, A.J.P.. Szanto, R. Garcia, P.J.Keely, P.. Mateos, E. Cukierman, A.. Miguel, Biomechanical remodeling of the microenvironment by stromal caveolin-1 164
favors tumor invasion and metastasis, Cell. 146 (2011) 148–163. https://doi.org/10.1016/j.cell.2011.05.040.
[226] G.S. Karagiannis, T. Poutahidis, S.E. Erdman, R. Kirsch, R.H. Riddell, E.P. Diamandis, Cancer-Associated Fibroblasts Drive the Progression of Metastasis through both Paracrine and Mechanical Pressure on Cancer Tissue, Mol. Cancer Res. 10 (2012) 1403–1418. https://doi.org/10.1158/1541-7786.MCR-12-0307.
[227] D. Liao, Y. Luo, D. Markowitz, R. Xiang, R.A. Reisfeld, Cancer associated fibroblasts promote tumor growth and metastasis by modulating the tumor immune microenvironment in a 4T1 murine breast cancer model, PLoS One. 4 (2009) e7965. https://doi.org/10.1371/journal.pone.0007965.
[228] S.L. Ham, S. Nasrollahi, K.N. Shah, A. Soltisz, S. Paruchuri, Y.H. Yun, G.D. Luker, A. Bishayee, H. Tavana, Phytochemicals potently inhibit migration of metastatic breast cancer cells, Integr. Biol. 7 (2015) 792–800. https://doi.org/10.1039/c5ib00121h.
[229] C. Neophytou, P. Boutsikos, P. Papageorgis, Molecular Mechanisms and Emerging Therapeutic Targets of Triple-Negative Breast Cancer Metastasis, Front. Oncol. 8 (2018). https://doi.org/10.3389/fonc.2018.00031.
[230] T. Masuda, M. Endo, Y. Yamamoto, H. Odagiri, T. Kadomatsu, T. Nakamura, H. Tanoue, H. Ito, M. Yugami, K. Miyata, J. Morinaga, H. Horiguchi, I. Motokawa, K. Terada, M. Suimye Morioka, I. Manabe, H. Iwase, H. Mizuta, Y. Oike, ANGPTL2 increases bone metastasis of breast cancer cells through enhancing CXCR4 signaling, Sci. Rep. 5 (2015) 9170. https://doi.org/10.1038/srep09170.
[231] H. Papatheodorou, A.. Papanastasiou, C. Sirinian, C. Scopa, H.P. Kalofonos, M. Leotsinidis, H. Papadaki, Expression patterns of SDF1/CXCR4 in human invasive breast carcinoma and adjacent normal stroma: correlation with tumor clinicopathological parameters, Pathol. - Res. Pract. 10 (2014) 662–667. https://doi.org/10.1016/j.prp.2014.06.015.
[232] A. Roemer, I. Luehr, Klein, A. Friedl, S. Sebens, F. Roesel, M. Bauer, Normal mammary fibroblasts induce reversion of the malignant phenotype in human primary breast cancer, Anticancer Res. 4 (2013) 1525–1536.
[233] T. Alkasalias, E. Flaberg, V. Kashuba, A. Alexeyenko, T. Pavlova, A. Savchenko, L. Szekely, G. Klein, H. Guven, Inhibition of tumor cell proliferation and motility by fibroblasts is both contact and soluble factor dependent, Proc. Natl. Acad. Sci. 111 (2014) 17188–17193. https://doi.org/10.1073/pnas.1419554111. 165
[234] S.F.G. Krens, J.H. Veldhuis, V. Barone, D. Capek, J.L. Maître, G.W. Brodland, C.P. Heisenberg, Interstitial fluid osmolarity modulates the action of differential tissue surface tension in progenitor cell segregation during gastrulation, Development. 144 (2017) 1798–1806. https://doi.org/10.1242/dev.144964.
[235] R.A. Foty, M.S. Steinberg, The differential adhesion hypothesis: A direct evaluation, Dev. Biol. 278 (2005) 255–263. https://doi.org/10.1016/j.ydbio.2004.11.012.
[236] C. Lan, A. Heindl, X. Huang, S. Xi, S. Banerjee, J. Liu, Y. Yuan, Quantitative histology analysis of the ovarian tumour microenvironment, Sci. Rep. 5 (2015) 1–12. https://doi.org/10.1038/srep16317.
[237] A. Ahmad, Breast Cancer Statistics: Recent Trends, in: Adv. Exp. Med. Biol., 2019. https://doi.org/10.1007/978-3-030-20301-6_1.
[238] I.A. Mayer, V.G. Abramson, B.D. Lehmann, J.A. Pietenpol, New Strategies for Triple-Negative Breast Cancer—Deciphering the Heterogeneity, Clin. Cancer Res. 20 (2014) 782–790. https://doi.org/10.1158/1078-0432.CCR- 13-0583.
[239] E.A. Rakha, M.E. El-Sayed, A.R. Green, A.H.S. Lee, J.F. Robertson, I.O. Ellis, Prognostic markers in triple-negative breast cancer, Cancer. 109 (2007) 25–32. https://doi.org/10.1002/cncr.22381.
[240] H. Gonçalves, M.R. Guerra, J.R. Duarte Cintra, V.A. Fayer, I.V. Brum, M.T. Bustamante Teixeira, Survival Study of Triple-Negative and Non–Triple- Negative Breast Cancer in a Brazilian Cohort, Clin. Med. Insights Oncol. 12 (2018) 117955491879056. https://doi.org/10.1177/1179554918790563.
[241] W.D. Foulkes, I.E. Smith, J.S. Reis-Filho, Triple-Negative Breast Cancer, N. Engl. J. Med. 363 (2010) 1938–1948. https://doi.org/10.1056/NEJMra1001389.
[242] P. Sharma, Biology and Management of Patients With Triple‐Negative Breast Cancer, Oncologist. 21 (2016) 1050–1062. https://doi.org/10.1634/theoncologist.2016-0067.
[243] L. Hutchinson, Breast cancer: TNBC: can we treat the untargetable?, Nat. Rev. Clin. Oncol. 11 (2014) 379. https://doi.org/10.1038/nrclinonc.2014.88.
[244] H. Masuda, K.A. Baggerly, Y. Wang, Y. Zhang, A.M. Gonzalez-Angulo, F. Meric-Bernstam, V. Valero, B.D. Lehmann, J.A. Pietenpol, G.N. Hortobagyi, W.F. Symmans, N.T. Ueno, Differential response to neoadjuvant chemotherapy among 7 triple-negative breast cancer 166
molecular subtypes, Clin. Cancer Res. 19 (2013) 5533–5540. https://doi.org/10.1158/1078-0432.CCR-13-0799.
[245] R. Rouzier, C.M. Perou, W.F. Symmans, N. Ibrahim, M. Cristofanilli, K. Anderson, K.R. Hess, J. Stec, M. Ayers, P. Wagner, P. Morandi, C. Fan, I. Rabiul, J.S. Ross, G.N. Hortobagyi, L. Pusztai, Breast cancer molecular subtypes respond differently to preoperative chemotherapy, Clin. Cancer Res. 11 (2005) 5678–5685. https://doi.org/10.1158/1078-0432.CCR-04- 2421.
[246] R. Kalluri, The biology and function of fibroblasts in cancer, Nat. Rev. Cancer. 16 (2016) 582–598. https://doi.org/10.1038/nrc.2016.73.
[247] T. Alkasalias, L. Moyano-Galceran, M. Arsenian-Henriksson, K. Lehti, Fibroblasts in the tumor microenvironment: Shield or spear?, Int. J. Mol. Sci. 19 (2018). https://doi.org/10.3390/ijms19051532.
[248] A.A. Seyhan, Lost in translation: the valley of death across preclinical and clinical divide – identification of problems and overcoming obstacles, Transl. Med. Commun. 4 (2019) 1–19. https://doi.org/10.1186/s41231-019- 0050-7.
[249] M. Singh, N. Ferrara, Modeling and predicting clinical efficacy for drugs targeting the tumor milieu, Nat. Biotechnol. 30 (2012) 648–657. https://doi.org/10.1038/nbt.2286.
[250] B.E. Miller, F.R. Miller, G.H. Heppner, Factors Affecting Growth and Drug Sensitivity of Mouse Mammary Tumor Lines in Collagen Gel Cultures, Cancer Res. 45 (1985) 4200–4205.
[251] X. Xu, M.C. Farach-Carson, X. Jia, Three-dimensional in vitro tumor models for cancer research and drug evaluation, Biotechnol. Adv. 32 (2014) 1256–1268. https://doi.org/10.1016/j.biotechadv.2014.07.009.
[252] R. MacArron, M.N. Banks, D. Bojanic, D.J. Burns, D.A. Cirovic, T. Garyantes, D.V.S. Green, R.P. Hertzberg, W.P. Janzen, J.W. Paslay, U. Schopfer, G.S. Sittampalam, Impact of high-throughput screening in biomedical research, Nat. Rev. Drug Discov. 10 (2011) 188–195. https://doi.org/10.1038/nrd3368.
[253] S. Singh, L.A. Ray, P.S. Thakuri, S. Tran, M.C. Konopka, G.D. Luker, H. Tavana, Organotypic breast tumor model elucidates dynamic remodeling of tumor microenvironment, Biomaterials. 238 (2020) 119853. https://doi.org/10.1016/j.biomaterials.2020.119853.
[254] J.A. Espinoza, S. Jabeen, R. Batra, E. Papaleo, V. Haakensen, V. 167
Timmermans Wielenga, M.L. Møller Talman, N. Brunner, A.L. Børresen- Dale, P. Gromov, Å. Helland, V.N. Kristensen, I. Gromova, Cytokine profiling of tumor interstitial fluid of the breast and its relationship with lymphocyte infiltration and clinicopathological characteristics, Oncoimmunology. 5 (2016) e1248015. https://doi.org/10.1080/2162402X.2016.1248015.
[255] S. Koeck, J. Kern, M. Zwierzina, G. Gamerith, E. Lorenz, S. Sopper, H. Zwierzina, A. Amann, The influence of stromal cells and tumor- microenvironment-derived cytokines and chemokines on CD3+CD8+ tumor infiltrating lymphocyte subpopulations, Oncoimmunology. 6 (2017) e1323617. https://doi.org/10.1080/2162402X.2017.1323617.
[256] E.J. Guggenheim, A. Khan, J. Pike, L. Chang, I. Lynch, J.Z. Rappoport, Comparison of confocal and super-resolution reflectance imaging of metal oxide nanoparticles, PLoS One. 11 (2016) e0159980. https://doi.org/10.1371/journal.pone.0159980.
[257] R. Joshi, P.S. Thakuri, J.C. Buchanan, J. Li, H. Tavana, Microprinted Stem Cell Niches Reveal Compounding Effect of Colony Size on Stromal Cells- Mediated Neural Differentiation, Adv. Healthc. Mater. 7 (2018) 1700832. https://doi.org/10.1002/adhm.201700832.
[258] R. Joshi, J.C. Buchanan, H. Tavana, Self-regulatory factors of embryonic stem cells in co-culture with stromal cells enhance neural differentiation, Integr. Biol. (United Kingdom). 9 (2017) 418–426. https://doi.org/10.1039/c7ib00038c.
[259] P.S. Thakuri, M. Gupta, R. Joshi, S. Singh, H. Tavana, Synergistic Inhibition of Kinase Pathways Overcomes Resistance of Colorectal Cancer Spheroids to Cyclic Targeted Therapies, ACS Pharmacol. Transl. Sci. 2 (2019) 275–284. https://doi.org/10.1021/acsptsci.9b00042.
[260] B. Pereira, S.F. Chin, O.M. Rueda, H.K.M. Vollan, E. Provenzano, H.A. Bardwell, M. Pugh, L. Jones, R. Russell, S.J. Sammut, D.W.Y. Tsui, B. Liu, S.J. Dawson, J. Abraham, H. Northen, J.F. Peden, A. Mukherjee, G. Turashvili, A.R. Green, S. McKinney, A. Oloumi, S. Shah, N. Rosenfeld, L. Murphy, D.R. Bentley, I.O. Ellis, A. Purushotham, S.E. Pinder, A.L. BØrresen-Dale, H.M. Earl, P.D. Pharoah, M.T. Ross, S. Aparicio, C. Caldas, The somatic mutation profiles of 2,433 breast cancers refines their genomic and transcriptomic landscapes, Nat. Commun. 7 (2016) 1–16. https://doi.org/10.1038/ncomms11479.
[261] E. Cerami, J. Gao, U. Dogrusoz, B.E. Gross, S.O. Sumer, B.A. Aksoy, A. Jacobsen, C.J. Byrne, M.L. Heuer, E. Larsson, Y. Antipin, B. Reva, A.P.
168
Goldberg, C. Sander, N. Schultz, A. Aksoy, A. Jacobsen, C.J. Byrne, M.L. Heuer, E. Larsson, Y. Antipin, B. Reva, A.P. Goldberg, C. Sander, N. Schultz, The cBio Cancer Genomics Portal: An open platform for exploring multidimensional cancer genomics data, Cancer Discov. 2 (2012) 401–404. https://doi.org/10.1158/2159-8290.CD-12-0095.
[262] J. Li, Y. Lu, R. Akbani, Z. Ju, P.L. Roebuck, W. Liu, J.Y. Yang, B.M. Broom, R.G.W. Verhaak, D.W. Kane, C. Wakefield, J.N. Weinstein, G.B. Mills, H. Liang, TCPA: A resource for cancer functional proteomics data, Nat. Methods. 10 (2013) 1046–1047. https://doi.org/10.1038/nmeth.2650.
[263] T. Kitamura, Y. Kato, D. Brownlie, D.Y.H. Soong, G. Sugano, N. Kippen, J. Li, D. Doughty-Shenton, N. Carragher, J.W. Pollard, Mammary tumor cells with high metastatic potential are hypersensitive to macrophage-derived HGF, Cancer Immunol. Res. 7 (2019) 2052–2064. https://doi.org/10.1158/2326-6066.CIR-19-0234.
[264] A. Subramanian, P. Tamayo, V.K. Mootha, S. Mukherjee, B.L. Ebert, M.A. Gillette, A. Paulovich, S.L. Pomeroy, T.R. Golub, E.S. Lander, J.P. Mesirov, Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 15545–15550. https://doi.org/10.1073/pnas.0506580102.
[265] J. Spangle, T. Roberts, Epigenetic regulation of RTK signaling, J. Mol. Med. (2017) 791–798. https://doi.org/10.1007/s00109-017-1546-0.
[266] J. Schlessinger, Receptor tyrosine kinases: Legacy of the first two decades, Cold Spring Harb. Perspect. Biol. 6 (2014) a008912. https://doi.org/10.1101/cshperspect.a008912.
[267] T. Brabletz, EMT and MET in Metastasis: Where Are the Cancer Stem Cells?, Cancer Cell. (2012) 699–701. https://doi.org/10.1016/j.ccr.2012.11.009.
[268] J.S. Duncan, M.C. Whittle, K. Nakamura, A.N. Abell, A.A. Midland, J.S. Zawistowski, N.L. Johnson, D.A. Granger, N.V. Jordan, D.B. Darr, J. Usary, P.F. Kuan, D.M. Smalley, B. Major, X. He, K.A. Hoadley, B. Zhou, N.E. Sharpless, C.M. Perou, W.Y. Kim, S.M. Gomez, X. Chen, J. Jin, S. V. Frye, H.S. Earp, L.M. Graves, G.L. Johnson, Dynamic reprogramming of the kinome in response to targeted MEK inhibition in triple-negative breast cancer, Cell. 149 (2012) 307–321. https://doi.org/10.1016/j.cell.2012.02.053.
[269] A.W. Tolcher, W. Peng, E. Calvo, Rational approaches for combination therapy strategies targeting the MAP kinase pathway in solid tumors, Mol. Cancer Ther. 17 (2018) 3–16. https://doi.org/10.1158/1535-7163.MCT-17- 169
0349.
[270] V.K. Gadiyaram, S. Kurian, J. Abraham, B. Ducatman, H. Hazard, G. Hobbs, L. Vona-Davis, Recurrence and survival after pulmonary metastasis in triple-negative breast cancer., J. Clin. Oncol. 28 (2010) 1131–1131. https://doi.org/10.1200/jco.2010.28.15_suppl.1131.
[271] A. Liberzon, C. Birger, H. Thorvaldsdóttir, M. Ghandi, J.P. Mesirov, P. Tamayo, The Molecular Signatures Database Hallmark Gene Set Collection, Cell Syst. 1 (2015) 417–425. https://doi.org/10.1016/j.cels.2015.12.004.
[272] N. Kanaji, M. Yokohira, Y. Nakano-Narusawa, N. Watanabe, K. Imaida, N. Kadowaki, S. Bandoh, Hepatocyte growth factor produced in lung fibroblasts enhances non-small cell lung cancer cell survival and tumor progression, Respir. Res. 18 (2017) 1–10. https://doi.org/10.1186/s12931- 017-0604-z.
[273] N.K. Altorki, G.J. Markowitz, D. Gao, J.L. Port, A. Saxena, B. Stiles, T. McGraw, V. Mittal, The lung microenvironment: an important regulator of tumour growth and metastasis, Nat. Rev. Cancer. 19 (2019) 9–31. https://doi.org/10.1038/s41568-018-0081-9.
[274] Y. Horimoto, U.M. Polanska, Y. Takahashi, A. Orimo, Emerging roles of the tumor-associated stroma in promoting tumor metastasis, Cell Adhes. Migr. 6 (2012) 193–203. https://doi.org/10.4161/cam.20631.
[275] J. Tchou, J. Conejo-Garcia, Targeting the Tumor Stroma as a Novel Treatment Strategy for Breast Cancer. Shifting from the Neoplastic Cell- Centric to a Stroma-Centric Paradigm, Adv. Pharmacol. 65 (2012) 45–61. https://doi.org/10.1016/B978-0-12-397927-8.00003-8.
[276] P. Chen, Q. Mo, B. Wang, D. Weng, P. Wu, G. Chen, Breast cancer associated fibroblasts promote MCF-7 invasion in vitro by secretion of HGF, J. Huazhong Univ. Sci. Technol. - Med. Sci. 32 (2012) 92–96. https://doi.org/10.1007/s11596-012-0016-8.
[277] A.B. Tuck, M. Park, E.E. Sterns, A. Boag, B.E. Elliott, Coexpression of hepatocyte growth factor and receptor (met) in human breast carcinoma, Am. J. Pathol. 148 (1996) 225–232.
[278] Y. Wang, A.C. Selden, N. Morgan, G.W.H. Stamp, H.J.F. Hodgson, Hepatocyte growth factor/scatter factor expression in human mammary epithelium, Am. J. Pathol. 144 (1994) 675–682.
[279] L. Jin, A. Fuchs, S.J. Schnitt, Y. Yao, A. Joseph, K. Lamszus, M. Park, I.D. 170
Goldberg, E.M. Rosen, Expression of scatter factor and c-met receptor in benign and malignant breast tissue, Cancer. 79 (1997) 749–760. https://doi.org/10.1002/(sici)1097-0142(19970215)79:4<749::aid- cncr12>3.0.co;2-%23.
[280] C. Ho-Yen, R.L. Bowen, S. Kermorgant, J.L. Jones, Comment on “High MET expression is an adverse prognostic factor in patients with triple- negative breast cancer,” Br. J. Cancer. 108 (2013) 2195–2196. https://doi.org/10.1038/bjc.2013.249.
[281] C.R. Graveel, J.D. DeGroot, Y. Su, J. Koeman, K. Dykema, S. Leung, J. Snider, S.R. Davies, P.J. Swiatek, S. Cottingham, M.A. Watson, M.J. Ellis, R.E. Sigler, K.A. Furge, G.F. Vande Woude, Met induces diverse mammary carcinomas in mice and is associated with human basal breast cancer, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 12909–12914. https://doi.org/10.1073/pnas.0810403106.
[282] M.G. Ponzo, R. Lesurf, S. Petkiewicz, F.P. O’Malley, D. Pinnaduwage, I.L. Andrulis, S.B. Bull, N. Chughtai, D. Zuo, M. Souleimanova, D. Germain, A. Omeroglu, R.D. Cardiff, M. Hallett, M. Park, Met induces mammary tumors with diverse histologies and is associated with poor outcome and human basal breast cancer, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 12903– 12908. https://doi.org/10.1073/pnas.0810402106.
[283] S. Garcia, J.P. Dalès, E. Charafe-Jauffret, S. Carpentier-Meunier, L. Andrac-Meyer, J. Jacquemier, C. Andonian, M.N. Lavaut, C. Allasia, P. Bonnier, C. Charpin, Poor prognosis in breast carcinomas correlates with increased expression of targetable CD146 and c-Met and with proteomic basal-like phenotype, Hum. Pathol. 38 (2007) 830–841. https://doi.org/10.1016/j.humpath.2006.11.015.
[284] R.A.D. Ghoussoub, D.A. Dillon, T. D’Aquila, E.B. Rimm, E.R. Fearon, D.L. Rimm, Expression of c-met is a strong independent prognostic factor in breast carcinoma, Cancer. 82 (1998) 1513–1520. https://doi.org/10.1002/(SICI)1097-0142(19980415)82:8<1513::AID- CNCR13>3.0.CO;2-7.
[285] P.S.H. Soon, E. Kim, C.K. Pon, A.J. Gill, K. Moore, A.J. Spillane, D.E. Benn, R.C. Baxter, Breast cancer-associated fibroblasts induce epithelial- to-mesenchymal transition in breast cancer cells, Endocr. Relat. Cancer. 20 (2013) 1–12. https://doi.org/10.1530/ERC-12-0227.
[286] M. Wang, J. Zhang, Y. Huang, S. Ji, G. Shao, S. Feng, D. Chen, K. Zhao, Z. Wang, A. Wu, Cancer-associated fibroblasts autophagy enhances progression of triple-negative breast cancer cells, Med. Sci. Monit. 23
171
(2017) 3904. https://doi.org/10.12659/MSM.902870.
[287] I. Jang, K.A. Beningo, Integrins, CAFs and mechanical forces in the progression of cancer, Cancers (Basel). 11 (2019) 721. https://doi.org/10.3390/cancers11050721.
[288] F. Demircioglu, J. Wang, J. Candido, A.S.H. Costa, P. Casado, B. de Luxan Delgado, L.E. Reynolds, J. Gomez-Escudero, E. Newport, V. Rajeeve, A.M. Baker, M. Roy-Luzarraga, T.A. Graham, J. Foster, Y. Wang, J.J. Campbell, R. Singh, P. Zhang, T.J. Schall, F.R. Balkwill, J. Sosabowski, P.R. Cutillas, C. Frezza, P. Sancho, K. Hodivala-Dilke, Cancer associated fibroblast FAK regulates malignant cell metabolism, Nat. Commun. 11 (2020) 1–18. https://doi.org/10.1038/s41467-020-15104- 3.
[289] H. Nakajima, Y. Ishikawa, M. Furuya, T. Sano, Y. Ohno, J. Horiguchi, T. Oyama, Protein expression, gene amplification, and mutational analysis of EGFR in triple-negative breast cancer, Breast Cancer. 21 (2014) 66–74. https://doi.org/10.1007/s12282-012-0354-1.
[290] M.M. Shao, F. Zhang, G. Meng, X.X. Wang, H. Xu, X.W. Yu, L.Y. Chen, G.M. Tse, Epidermal growth factor receptor gene amplification and protein overexpression in basal-like carcinoma of the breast, Histopathology. 59 (2011) 264–273. https://doi.org/10.1111/j.1365-2559.2011.03921.x.
[291] V. Martin, F. Botta, E. Zanellato, F. Molinari, S. Crippa, L. Mazzucchelli, M. Frattini, Molecular characterization of EGFR and EGFR-downstream pathways in triple negative breast carcinomas with basal like features, Histol. Histopathol. 27 (2012) 785–792. https://doi.org/10.14670/HH- 27.785.
[292] H.S. Park, M.H. Jang, E.J. Kim, H.J. Kim, H.J. Lee, Y.J. Kim, J.H. Kim, E. Kang, S.W. Kim, I.A. Kim, S.Y. Park, High EGFR gene copy number predicts poor outcome in triple-negative breast cancer, Mod. Pathol. 27 (2014) 1212–1222. https://doi.org/10.1038/modpathol.2013.251.
[293] R. Ali, M.K. Wendt, The paradoxical functions of EGFR during breast cancer progression, Signal Transduct. Target. Ther. 2 (2017) 1–7. https://doi.org/10.1038/sigtrans.2016.42.
[294] M. Pein, J. Insua-Rodríguez, T. Hongu, A. Riedel, J. Meier, L. Wiedmann, K. Decker, M.A.G. Essers, H.P. Sinn, S. Spaich, M. Sütterlin, A. Schneeweiss, A. Trumpp, T. Oskarsson, Metastasis-initiating cells induce and exploit a fibroblast niche to fuel malignant colonization of the lungs, Nat. Commun. 11 (2020) 1–18. https://doi.org/10.1038/s41467-020-15188- x. 172
[295] Y. Del Pozo Martin, D. Park, A. Ramachandran, L. Ombrato, F. Calvo, P. Chakravarty, B. Spencer-Dene, S. Derzsi, C.S. Hill, E. Sahai, I. Malanchi, Mesenchymal Cancer Cell-Stroma Crosstalk Promotes Niche Activation, Epithelial Reversion, and Metastatic Colonization, Cell Rep. 13 (2015) 2456–2469. https://doi.org/10.1016/j.celrep.2015.11.025.
[296] M.I. Gallego, B. Bierie, L. Hennighausen, Targeted expression of HGF/SF in mouse mammary epithelium leads to metastatic adenosquamous carcinomas through the activation of multiple signal transduction pathways, Oncogene. 22 (2003) 8498–8508. https://doi.org/10.1038/sj.onc.1207063.
[297] W.J. Lee, W.K. Chen, C.J. Wang, W.L. Lin, T.H. Tseng, Apigenin inhibits HGF-promoted invasive growth and metastasis involving blocking PI3K/Akt pathway and β4 integrin function in MDA-MB-231 breast cancer cells, Toxicol. Appl. Pharmacol. 226 (2008) 178–191. https://doi.org/10.1016/j.taap.2007.09.013.
[298] R. Harisi, A. Jeney, Extracellular matrix as target for antitumor therapy, Onco. Targets. Ther. 8 (2015) 1387–1398. https://doi.org/10.2147/OTT.S48883.
[299] C. Walker, E. Mojares, A. Del Río Hernández, Role of extracellular matrix in development and cancer progression, Int. J. Mol. Sci. 19 (2018). https://doi.org/10.3390/ijms19103028.
[300] M. Rubart, Two-photon microscopy of cells and tissue, Circ. Res. 95 (2004) 1154–1166. https://doi.org/10.1161/01.RES.0000150593.30324.42.
[301] D. Kobat, N.G. Horton, C. Xu, In vivo two-photon microscopy to 1.6-mm depth in mouse cortex, J. Biomed. Opt. 16 (2011) 106014. https://doi.org/10.1117/1.3646209.
[302] X. Liu, V. Ory, S. Chapman, H. Yuan, C. Albanese, B. Kallakury, O.A. Timofeeva, C. Nealon, A. Dakic, V. Simic, B.R. Haddad, J.S. Rhim, A. Dritschilo, A. Riegel, A. McBride, R. Schlegel, ROCK inhibitor and feeder cells induce the conditional reprogramming of epithelial cells, Am. J. Pathol. 180 (2012) 599–607. https://doi.org/10.1016/j.ajpath.2011.10.036.
[303] H. Yuan, S. Myers, J. Wang, D. Zhou, J.A. Woo, B. Kallakury, A. Ju, M. Bazylewicz, Y.M. Carter, C. Albanese, N. Grant, A. Shad, A. Dritschilo, X. Liu, R. Schlegel, Use of reprogrammed cells to identify therapy for respiratory papillomatosis, N. Engl. J. Med. 367 (2012) 1220–1227. https://doi.org/10.1056/NEJMoa1203055.
173
APPENDIX
Figure A-1: MET RTK activated by TNBC-CAFs interaction. Phospho-RTK arrays of (a) MDA-MB-231 and (b) SUM159 cells stimulated with conditioned media of CAF-1 and CAF-2 monocultures showing pixel integrated densities representing activities of several RTKs from treatments normalized with the respective vehicle control and represented as activity fold change. Error bars represent standard errors from a mean value. Data represent two separate experiments.
174
175
176