CANCER-CELL INTRINSIC MECHANSIMS OF MEDIATING RESPONSE TO THERAPEUTICS IN NSCLC
BY
EMILY KAY WAGNER
B.S., Rochester Institute of Technology, 2018
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment
of the requirements for the degree of Master of Science
Pharmacology Training Program 2020
ii
This thesis for the Master of Science degree by Emily Kay Wagner
has been approved for the Pharmacology Training Program
by
Lynn Heasley, Chair Philip Owens Rebecca Schweppe Raphael Nemenoff, Advisor
Date: December 11, 2020
iii
Wagner, Emily Kay (M.S., Biomedical Sciences and Biotechnology)
Cancer-cell intrinsic mechanisms of mediating response to therapeutics in NSCLC Thesis directed by Professor Raphael A. Nemenoff
ABSTRACT
Lung cancer is the leading cause of cancer deaths in the United States, with a 5-year survival rate of approximately 20%. Non-small cell lung cancer (NSCLC) comprises 85% of lung cancer diagnoses. Clinically available therapies used to treat NSCLC include targeted therapies towards driving oncogenic mutations, and immunotherapy. Unfortunately, patients on targeted therapies often acquire new mutations and relapse; additionally, only 20% of unselected patients respond well to immunotherapy treatment. Thus, there is a need to determine new therapeutic strategies in treating NSCLC.
The tumor microenvironment (TME) is the cellular milieu comprised of tumor cells and immune cells, which can work to promote or inhibit tumor growth. The efficacy of
immunotherapies has been found to be reliant on the presence of T cells within the TME. Recent publications have highlighted the importance of surface expression of Major Histocompatibility Complex II (MHCII) on cancer cells in mediating T cell recruitment and enhancing response to immunotherapies. The goal of this project was to determine the effects of clinically available small molecule inhibitors of MAP kinase signaling and histone deacetylase (HDAC) signaling on MHC II expression and T cell recruitment. Murine and human non-small cell lung cancer
(NSCLC) cell lines containing various activating mutations were treated with inhibitors of MAP kinase and HDAC signaling to determine the effects on MHCII and complement protein C3 expression.
C-C Motif Chemokine Ligand 2 (CCL2) is a chemoattractant of monocytes and other
iv
myeloid-derived immune cells. In a previously published paper from our lab, CCL2 was
identified as a possible mediator of resistance to immunotherapy. Using cancer cells containing a CCL2 knockdown, mice were treated with anti-PD-1 therapy to determine the effect on tumor growth. Overall, these experiments seek to determine how cancer cells may modulate the TME to enhance response to therapeutics, particularly immunotherapy.
The form and content of this abstract are approved. I recommend its publication.
Approved: Raphael A. Nemenoff
v
I would like to dedicate this thesis in honor of the late Dr. David A. Lawlor, an RIT professor who inspired my love for immunology and a passion for translational research.
vi
ACKNOWLEDGEMENTS
This thesis was supported by the T32 Training Grant in Pharmacology (NIH NRSA
T32GM007635-41) and the T32 Training Grant for Training in Translational Research of Lung, Head and Neck Cancer (NIH NRSA T32CA174648). I would like to thank Rafe for taking me into his lab for the past few years. Thanks to Bonnie Bullock, Amber Johnson, and Emily Kleczko for being great lab mates, mentors, and friends; without you all to talk to, I wouldn’t be where I am today. Thank you to Dr. Port, Shanelle Felder, and the rest of the Pharmacology Training Program. Finally, thank you to all of the mice from under the approved IACUC protocol 148, your sacrifice was greatly appreciated.
vii
TABLE OF CONTENTS CHAPTER
I: INTRODUCTION………1
II: EFFECTS OF HDAC INHIBITION ON EML4/ALK FUSION CELL
LINES……….……….6
III: EFFECTS OF HDAC ON HUMAN AND MURINE INHIBITION ON
KRAS-MUTANT CELL LINES………13
IV: CCL2 KNOCKDOWN IN LLC CELLS ENHANCES RESPONSE TO
IMMUNOTHERAPY………...18
V: CONCLUSIONS AND FUTURE DIRECTIONS……….28
REFERENCES………30
1 CHAPTER I INTRODUCTION Lung Cancer
Lung cancer is the leading cause of cancer deaths in the United States, with a 5-year survival rate of less than 20%1. Non-small cell lung cancer (NSCLC) is a heterogeneous disease that accounts for 85% of the cases, driven by a number of oncogenic drivers2. Of these, there are two in particular of interest in this thesis: KRAS mutations and EML4/ALK fusions. In the past 10 to 15 years, there have been two crucial advancements in the treatment of lung cancer. The first of these is the development of targeted therapies, which are effective in tumors driven by activating mutations and rearrangement of tyrosine kinases. However, while the majority of patients respond to these therapies, many have only a partial response and all patients eventually develop resistance. The second is the development of immune checkpoint therapies, such as anti-PD-1 and anti-PD-L1. Unfortunately, these are effective in only ~20% of unselected patients3-5. While both of these therapies are effective to an extent, partial responses or development of resistance to them are common, highlighting the need for new therapeutic targets.
Immunotherapy
A recent advancement in the treatment of cancer is the development of monoclonal antibodies that target immune checkpoint proteins expressed on both T cells and tumor cells.
Programmed death 1 (PD-1) is found on the surface of T cells, and when bound to programmed death ligand 1 (PD-L1) found on the surface of tumor cells and other cells, turns off T cell function and renders them in an exhaustive state8. Tumor cells that hijack this pathway can then evade immune destruction, characterized as one of the hallmarks of cancer9. Monoclonal
2
antibodies are used as therapeutics to prevent interaction of PD-1 and PD-L1, allowing T cells to remain activated and participate in active tumor cell killing. These therapies are highly effective in approximately 20% of unselected patients, emphasizing the need to potentially induce
sensitivity in the remaining 80%10.
Previously, the expression of cancer cell-specific MHCII (csMHCII) was found to correlate with enhanced response to immunotherapy in melanoma11. Additionally, the Nemenoff lab has demonstrated that an increase in csMHCII results in an enhanced response to anti-PD-1 therapies in NSCLC12. Induction of MHCII is mediated by inflammatory cytokines, such as IFN-
, in non-antigen presenting cells (such as cancer cells)13. Thus, there is evidence that supports the idea that inducing surface MHCII expression in patients may render tumors more sensitive to immunotherapies, resulting in better clinical outcomes.
Histone Deacetylases (HDACs) in Cancer
HDACs are enzymes that are responsible for the removal of -amino acetyl groups on lysine residues of histones. Their main function is to repress transcription by removing these acetyl groups, resulting in a more condensed chromatin that discourages transcription factor binding. There are 18 HDACs in humans which are divided into Classes I-IV. Class I HDACs consist of HDAC1, 2, 3, and 8; Class IIa HDACs consist of HDAC4, 5, 7, and 9; Class IIb HDACs consist of HDAC6, 10, and 116. Dysregulation of these enzymes can negatively impact gene regulation, and is often found in a variety of cancers including NSCLC, ovarian and gastric cancers, and multiple myeloma7. Currently, there are a number of clinically available HDAC inhibitors (HDACi) that work by inhibiting the deacetylation activity of the proteins, allowing chromatin to remain accessible and promote genes with anti-cancer activity. These include pro- apoptotic genes that lead to cell cycle arrest and death. Trichostatin A (TSA) is a pan-HDAC
3
inhibitor used in many experiments in this thesis. It targets both Class I and Class II HDACs, giving it the potential to elicit a number of responses in cells. HDACi are used mainly in the treatment of hematological cancers as single agent therapy7,8.
Combination Therapies with HDACi and Immunotherapy
While HDACi are effective as single agent therapy for hematological cancers, these drugs exhibit diminished efficacy when treating solid tumors. This phenomenon is most likely due to their short half-lives and poor pharmacokinetic properties8. As a result, the drug is metabolized and cleared from the body before it can enter the tumor and become effective.
However, combination therapy with chemotherapy, radiation and immunotherapy is currently being pursued in a number of clinical trials14-17. There is also growing evidence that these
combination therapies, particularly with immunotherapy, are successful at activating the immune system and resulting in tumor cell killing. One study demonstrated that treatment with HDACi can enhance the immunotherapy response of melanoma cells through increased cytokine secretion, induced expression of MHCA, an MHC Class I (MHCI) molecule, and a decrease in surface PD-L1 expression on tumor cells 18. Another study showed that treatment of lung cancer cell lines with HDACi vorinostat or romidepsin resulted in increased secretion of T cell
chemokines such as CCL5, CXCL9, and CXCL10. In addition, treatment with romidepsin in a number of tumor models enhanced response to PD-1 blockade19. HDACi not only impact tumor cells but immune cells in the tumor microenvironment as well. HDAC3 has been previously identified as a negative regulator of CD8+ T cell cytotoxicity; inhibiting it pharmacologically or via knockout results in increased expression of T cell effector genes such as granzyme B and perforin, and pro-inflammatory cytokines/chemokines such as CCL3, 4, and 5,
IFN-γ and TNF-α20. Treatment with entinostat, a selective Class I HDAC inhibitor, results in
4
increased natural killer (NK) cell function via upregulation of NKG2D, an NK cell activating receptor; additionally, IFN- production is increased, further contributing to an inflamed tumor microenvironment to mediate tumor cell killing21.
The Role of Complement in Cancer
The complement pathway is part of the innate immune system, named for its function to
“complement” the activity of phagocytic cells and immunoglobulin in clearing damaged cells and pathogens, promoting inflammation through recruitment of immune cells, and its intrinsic ability to attack pathogen membranes. Complement proteins are produced by the liver and released to circulate around the body to interact with cells via receptors expressed on cell
surfaces22. While complement is known for its antimicrobial ability, it has also been studied as a mediator of cancer progression. Complement component 3 (C3) is activated through one of three distinct pathways: classical, alternative, or lectin. At the end of all of these pathways, C3 is converted into C3a and C3b, which go on to participate further in complement signaling.
Importantly, C3a is an anaphylatoxin, which promotes cell migration, activation, and
inflammation. C3a binds to its cognate receptor C3aR, a G protein coupled receptor found on myeloid cells, such as dendritic cells, monocytes and macrophages, and neutrophils23. This binding initiates an oxidative burst from these cells that supports inflammation. C3a can also bind its receptor found on T cells, thus influencing their proliferation, expansion, differentiation, and viability. Complement has been found to have both pro- and anti-tumorigenic properties, highlighting the complex nature of these proteins in the context of cancer. Pro-tumorigenic functions of complement protein C3 include the recruitment of myeloid-derived suppressor cells (MDSCs), induction of CC-chemokine ligand 2 (CCL2) production that recruits M2-like
macrophages, and recruitment of tumor-promoting neutrophils. On the other hand, anti-
5
tumorigenic functions of C3 include induction of IFN- production, recruitment of tumor-
eliminating M1-like macrophages and NK cells, and enhanced recruitment of CD4+ and CD8+ T cells22,24.
Currently, inhibitors are being developed towards complement protein receptors and are now considered to be a new class of immune checkpoints that may be targeted like PD-1/PD-L1.
Monoclonal antibodies (mAbs) that target complement receptors alone or in combination with anti-PD-1/PD-L1 mAbs have shown efficacy in mouse models at reducing tumor burden and are currently in clinical trials22.
6 CHAPTER II
EFFECTS OF HDAC INHIBITON ON EML4/ALK FUSION CELL LINES Introduction
Previous publications have highlighted the importance of MHCII expression on cancer cells and have noted the benefits of improved survival and enhanced response to therapeutics. A study done by Neuwelt et al. demonstrated that inhibition of MAP kinase signaling through treatment with trametinib (MEK 1/2 inhibitor) and inhibition of HDAC signaling with TSA results in increased surface expression of MHCII in LLC cells25. From this, we hypothesized that, in an EML4/ALK driven lung cancer model, epigenetic modifications occurring due to HDAC inhibition lead to an increased sensitivity to IFN-γ, and thus a subsequent increase in CIITA expression. In addition, we explore the effects of HDAC inhibition in combination with the targeted agent alectinib.
Methods
EA1 and EA2 cell lines containing an activating EML4/ALK fusion were generated using the CRISPR-Cas9 system in a murine model26. The lines were cultured in RPMI with L-
glutamine (Corning, Inc.), containing 10% FBS and 1% penicillin/streptomycin at 37°C. The cells were seeded into 6-well plates at 100,000 cells per well for 18- and 24-hour time point, or 75,000 cells per well for 48-hour time point and incubated overnight. The cells were dosed in triplicate in the following scheme: DMSO/PBS control, IFN-γ alone, TSA alone, alectinib alone, combination of alectinib and TSA, or combination of IFN-γ and TSA. 24 hours after seeding, the appropriate wells were pre-treated with DMSO control or trichostatin A (TSA, 100nM), then incubated overnight. The following day, the appropriate wells were treated with 1X PBS control, alectinib (10nM), or IFN-γ (10ng/mL) and plates were incubated for 18, 24, or 48 hours.
7
Following incubation, the cells were washed with 1X PBS three times, and RNA extraction was performed using the RNeasy kit (Qiagen). RNA concentration was measured, and cDNA was synthesized using qScript XLT cDNA SuperMix (QuantaBio). Real Time PCR (RT-PCR) was performed on a Bio-Rad iCycler (Bio-Rad) using SYBR Green (Applied Biosystems). Primers were used to detect expression of complement protein C3 (5’- CCAGCTCCCCATTAGCTCTG- 3’ and 5’- GCACTTGCCTCTTTAGGAAGTC-3’), master regulator of MHC Class II gene CIITA (5’-TGCGTGTGATGGATGTCCAG-3’ and 5’-CCAAAGGGGATAGTGGGTGTC-3’), CXCL9 (5’-GAGCAGTGTGGAGTTCGAGG-3’ and 5’-TCCGGATCTAGGCAGGTTTG-3’), CXCL10 (5’-GGATGGCTGTCCTAGCTCTG-3’ and 5’-TGAGCTAGGGAGGACAAGGA- 3’), and housekeeping gene β-actin as a control (5’-GGCTGTATTCCCCTCCATCG-3’ and 5’- CCAGTTGGTAACAATGCCATG-3’).
Results The Effects of HDAC Inhibition on CIITA Expression
Pre-treatment of EA1 and EA2 cells with TSA, followed by treatment with IFN-γ induced expression of CIITA, CXCL9 and CXCL10 (Figures 2.1 and 2.2). In both EA1 and EA2 cell lines, IFN-γ alone was sufficient in inducing CIITA expression; in EA1 cells, an additive effect was observed when cells were pretreated with TSA prior to IFN-γ stimulation. This finding did not hold true in the EA2 cell line.
The Effects of HDAC Inhibition on Chemokine Expression
It has been previously found that cells with increased CIITA expression within the TME recruit more T cells12. With this, we decided to look at expression of CXCL9 and CXCL10, both known T cell recruiting chemokines. In both EA1 and EA2 cell lines, IFN-γ was sufficient in
8
inducing CXCL9 expression, but not CXCL10. Additionally, CXCL10 expression increased in response to TSA + Alectinib treatment, and TSA + IFN-γ treatment in both cell lines.
Effects of MEK Inhibition on Complement and CIITA Expression
Finally, we looked to determine the effects of MEK1/2 inhibition alone or in combination with IFN-γ to observe if the effects were similar to those when cells are treated with the HDAC inhibitor TSA. At baseline, EA1 cells express C3, and this effect was amplified when treated with a combination of IFN-γ + trametinib. EA1 cells also greatly induced expression of CIITA when treated with IFN-γ and trametinib. This was similar to the results seen when the cells were treated with IFN-γ and TSA.
Discussion and Future Directions
These experiments were conducted to test the hypothesis that, in an EML4/ALK driven model of lung cancer, HDAC inhibition and subsequent epigenetic modifications would lead to an increase in IFN- sensitivity and induction of MHCII via CIITA expression. In EA1 cells, IFN- and TSA induce CIITA expression, but not in EA2 cells. The combination also results in an increase in CXCL9 and CXCL10 expression. Trametinib treatment in EA1 cells recapitulated the results seen with TSA treatment, suggesting an overlap in signaling and/or function between MAP kinases and HDACs. Due to the similarity in response, combination of HDACi with alectinib or other targeted agents may result in a synergistic response in patients. The increase in MHCII expression and recruiting chemokines could result in greater T cell recruitment and decreased tumor burden.
TSA is a pan-HDAC inhibitor that targets both Class I and Class II HDACs. In the future, it would be imperative to identify single HDAC enzymes within those classes to selectively inhibit. This can be addressed by using drugs that target specific HDACs, or by performing a
9
knockdown or knockout of the enzyme of interest. Each of these methods would provide different information in terms of the degree of inhibition or loss. When treating with specific HDAC inhibitors, the protein would lose enzymatic function, while maintaining expression in the cells. A knockdown using shRNA would diminish expression to an extent, but likely leave some endogenous expression. A knockout using CRISPR-Cas9 technology would remove the HDAC completely from the cell’s genome, leaving the cell to compensate for the loss of the enzyme.
In order to determine the exact effect that the drug treatments have on the secretion of CXCL9 and CXCL10, an ELISA needs to be performed. This would directly measure the levels of secreted chemokine in the media, as opposed to only using message levels that may not be translated into protein readout.
10
Figure 2.1: Treatment of EA1 and EA2 cells with IFN-γ and TSA induces CIITA mRNA expression. (A, B): EA1 and EA2 cells were treated with combinations of IFN-γ (10 ng/mL), TSA (100 nM) and/or alectinib (10 nM). CIITA mRNA levels were measured using RT-PCR and were normalized to β-actin expression levels. N=1.
11
Figure 2.2: Treatment of EA1 and EA2 cells with IFN-γ, TSA or alectinib induces mRNA expression of CXCL9 and CXCL10. (A-D): EA1 and EA2 cells were treated with combinations of IFN-γ (10 ng/mL), TSA (100 nM) and/or alectinib (10 nM). CXCL9 and CXCL10 mRNA levels were measured using RT-PCR and were normalized to β-actin expression levels. N=1.
12
Figure 2.3: Treatment of EA1 cells with IFN-γ and trametinib induces expression of
complement protein C3 expression and CIITA. EA1 cells were treated alone or in combination with IFN-γ (10 ng/mL) and/or trametinib (10 nM). C3 and CIITA mRNA levels were measured using RT-PCR and were normalized to β-actin expression levels. N=1.
13 CHAPTER III
EFFECTS OF HDAC INHIBITION ON HUMAN AND MURINE KRAS-MUTANT CELL LINES
Introduction
A recent advancement in treating cancer has been the development of inhibitors that target mutations in KRAS proteins, particularly in regard to the G12C mutation. KRAS mutations are the most commonly found mutations in NSCLC; approximately 25% of patients have a mutation in KRAS. Smokers are more likely to develop these compared to non-smokers (30% to 10%, respectively). KRAS is an integral part of the MAP kinase pathway and signals downstream to RAF-MEK-ERK. Thus, we hypothesized that inhibiting KRAS with the targeted agent AMG510 will elicit the same effects as downstream inhibition of MEK1/2 using
trametinib, resulting in increased CIITA expression. Combination of AMG510 with TSA and IFN- will help us expand our understanding of how these treatments affect MHCII expression in a KRAS-driven lung cancer model.
Methods
Human KRAS-mutant cells lines (H358 and H1573) were cultured in RPMI with L- glutamine (Corning, Inc.), containing 10% FBS and 1% penicillin/streptomycin at 37°C. Murine Lewis Lung Carcinoma (LLC) cells were cultured in DMEM with 4.5 g/L glucose and L-
glutamine without sodium pyruvate (Corning, Inc.), containing 10% FBS and 1%
penicillin/streptomycin at 37°C. The cells were seeded at 300,000 cells per well in 6-well plates and incubated overnight. The following day, H358 and LLC cells were dosed in triplicate in the following scheme: DMSO/PBS control, IFN-γ alone (10ng/mL), AMG510 alone (10nM), or combination of IFN-γ and AMG-510; the plates were then incubated for 48 hours.
14
H1573 cells were seeded at 250,000 cells per well in 6-well plates and incubated overnight. The cells were dosed in triplicate in the following scheme: DMSO/PBS, IFN-γ alone, TSA alone, or in combination with IFN-γ, trametinib alone or in combination with IFN-γ, or AMG510 alone or in combination with IFN-γ. 24 hours after seeding, the appropriate wells were treated DMSO controls or TSA (100nM) and incubated overnight. 24 hours after TSA pretreatment, the remaining wells were treated with PBS control, IFN- γ (10ng/mL), trametinib (10nM), or alectinib (10nM), and were incubated overnight. Following drug treatment and incubation, all plates were washed with 1X PBS three times, and RNA extraction was performed using the RNeasy kit (Qiagen). RNA concentration was measured, and cDNA was synthesized using qScript XLT cDNA SuperMix (QuantaBio). Real Time PCR (RT-PCR) was performed on a Bio- Rad iCycler (Bio-Rad) using SYBR Green (Applied Biosystems). Primers were used to detect the expression of human GAPDH as a housekeeping gene
(5’- GTCAACGGATTTGGTCGTATTG -3’ and 5’- TGGAAGATGGTGATGGGATTT -3’) and human CIITA (5’- CCTGGAGCTTCTTAACAGCGA -3’ and 5’-
TGTGTCGGGTTCTGAGTAGAG -3’). The murine primers used for detecting expression of CIITA and β-actin can be found in the Methods section of Chapter II.
Results
Inhibition of KRAS G12C trends to an increase in CIITA Expression
Human H358 and murine LLC cells were treated with controls, or various combinations of IFN-γ and the specific KRAS G12C inhibitor AMG510 (Figure 3.1). In H358 cells, treatment with AMG510 resulted in an increase in CIITA expression; this effect was not amplified by the addition of IFN-γ. In LLC cells, AMG510 treatment resulted in a minimal increase in CIITA
15
expression, and the addition of IFN-γ had no effect. This was to be expected, as LLC cells are traditionally unresponsive to IFN-γ stimulation.
Treatment of H1573 cells with panel of drugs reveals differential response
H1573 cells containing a KRAS G12A mutation were treated with a panel of drugs to determine how each impacts the expression of CIITA (Figure 3.2). When dosed alone, TSA, trametinib and AMG510 do not induce CIITA expression. However, stimulation with IFN-γ in addition to treatment with AMG510 and TSA resulted in a slight increase in CIITA expression;
this is most likely due to the IFN-γ and not the drug treatments.
Discussion and Future Directions
These experiments were conducted to test the hypothesis that, in a KRAS-driven model of lung cancer, HDAC inhibition in combination with KRAS inhibition would result in increased sensitivity to IFN- and subsequent expression of CIITA. In both LLC and H358 cells, AMG510 induced expression of CIITA, but this effect was not compounded by IFN- treatment. In H1573 cells, the combination of trametinib and IFN- proved to be most effective. These results
demonstrate the differential effects of the same drugs used on various cell lines.
After these experiments were performed, it was discovered that LLC cells harbor an activating NRAS mutation27. Most likely, this mutation will compensate for the inhibition of KRAS using AMG510, and alter the response to the drug. Knocking down or knocking out this NRAS mutation will be crucial to observing the true effect that AMG510 has on these cells. The most effective method would be to use the CRISPR-Cas9 system to achieve this.
It will also be important that these experiments are continued in more cell lines to
observe differential effects. There are other KRAS mutations to investigate, including G12V and G12D, which are relevant in NSCLC. While there are no specific inhibitors available to target
16
these mutations as there are with G12C, it would remain beneficial to determine how TSA and trametinib treatments effect expression of CIITA and complement proteins.
17
Figure 3.1: Treatment of H358 and LLC cells with IFN-γ and targeted G12C inhibitor.
KRAS-mutant H358 and LLC cells were treated alone or in combination with IFN-γ (10 ng/mL) and/or AMG510 (10nM) for 48 hours. CIITA mRNA levels were measured using RT-PCR and were normalized to β-actin (LLC) or GAPDH (H358) expression levels. N=1.
Figure 3.2: Treatment of H1573 cells with IFN-γ and panel of drugs to determine changes in CIITA expression. KRAS-mutant H1573 cells were treated with alone or in combination with IFN-γ (10 ng/mL), TSA (100 nM), trametinib (10 nM), or AMG510 (10nM) for 48 hours. CIITA mRNA levels were measured using RT-PCR and were normalized to GAPDH. N=1.
18 CHAPTER IV
CCL2 KNOCKDOWN IN LLC CELL ENHANCES RESPONSE TO IMMUNOTHERAPY
Introduction
Previous work done in the Nemenoff lab by Bonnie Bullock showed that secretion of chemokine CCL2 in the TME may play a role in response to immunotherapy28. CCL2 is a known recruiter of monocytes and macrophages, particularly MDSCs, which contributes to
immunotherapy resistance. LLC cells, by RNAseq, express high levels of CCL2. With this, we hypothesize that knockdown of CCL2 in LLC cells will result in decreased recruitment of immature monocytes into the lung TME and result in an enhanced response to immunotherapy.
In addition, CMT167 cells express low levels of CCL2; as a complementary approach to
experiments with LLC cells, CCL2 was overexpressed in CMT167 cells to observe the effect on tumor growth. We hypothesized that increased expression of CCL2 would result in resistance to immunotherapy and increased tumor burden.
Methods Preparation of CCL2 Knockdown Cells
Murine shRNA constructs were obtained from Sigma-Aldrich via the University of Colorado Functional Genomics Shared Resource (TRC1) and the following cell lines were generated using them: Nontargeting control (SHC001V) and shRNA targeting CCL2 (TRCN0000034470). HEK293T cells were used to generate lentiviral particles containing desired shRNA vectors and helper plasmids. Viral supernatant was collected at 24 and 48 hours after transfection. Prior to transduction with viral media, LLC cells were treated with polybrene for 1 hour; polybrene was also added to viral supernatant and filtered through a 0.45 um filter
19
before media was added to LLC cells. After 10 days, stably expressing cells were selected using puromycin (2 μg/mL). Cells were pooled and the extent of knockdown was measured by RT- PCR for mRNA and by western blot for protein levels; levels were compared to parental and nontargeting cell lines for controls.
Preparation of CCL2 Overexpression Cells
ORF expression clones for CCL2 were obtained from GeneCopoeia (CCL2: EX- Mm05119-Lv152; EGFP: EX-EGFP-Lv152). Lentiviral transduction was performed on CMT167 cells the same as above in the “Preparation of CCL2 Knockdown Cells” section. 10 days after transduction, stably expressing cells were selected using hygromycin (2 μg/mL). Cells were pooled, and the extent of overexpression was measured by RT-PCR for mRNA, where the levels were compared to EGFP cell line.
Injection of Cells into Mice
Wild-type C57BL/6J mice were obtained from Jackson Laboratory. All mice were bred and maintained in the Center for Comparative Medicine at the University of Colorado Anschutz Medical Campus. Experiments were performed on 8-12-week old male mice. Nontargeting control and CCL2 knockdown cells were cultured in DMEM with 4.5 g/L glucose and L- glutamine without sodium pyruvate (Corning, Inc.), containing 10% FBS. On the day of
injection, tumor cells were suspended in 1.35 mg/mL Matrigel in Hank’s buffered saline solution at 1x105 cells in 40 μl per injection. The orthotopic model utilized was developed by a previous member of the Nemenoff laboratory29. An incision is made on the left lateral axillary line at the xiphoid process level, and subcutaneous fat is removed to reveal the ribs. Cells are injected into the left lung lobe through the rib cage with a 30-gauge needle. Tumors were then established for 7 days, followed by intraperitoneal treatment with control IgG2a isotype control antibody
20
(BioXCell #BE0089 Clone 2A3) or anti-PD-1 antibody (BioXCell #BE0146 Clone RMP1-14) for 2 weeks at concentrations of 8-10 mg/kg. After treatment was completed, the mice were euthanized and the lungs and heart were perfused with PBS/heparin. Tumor-bearing left lung lobes were removed.
Flow Cytometry Preparation and Performance
Tumor-bearing left lung lobes were pooled and mechanically and chemically dissociated into a single-cell suspension as previously described29. Cells were then stained with either the T cell phenotypic panel or myeloid cell phenotypic panel (Table 1). Cells were prepared using the FoxP3 Staining Buffer set (eBioscience #00-4970-93). The VersaComp Antibody Capture Bead Kit (Beckman Coulter #B22804) was used as a control for antibody stains. Samples were run on the Gallios system (Beckman Coulter) at the University of Colorado Cancer Center Flow
Cytometry Core. All data were analyzed using Kaluza software (Beckman Coulter). Gating strategies involved excluding dead cells using a cell viability dye, and doublet cells and other debris using light scatter. Single-stain controls were also used to determine gating strategy as previously described29,30.
Results
CCL2 Knockdown in LLC cells results in decreased tumor volume
LLC cells containing non-targeting shRNA (LLC-NT) or shRNA targeting CCL2 (LLC- sh70) were injected into mice orthotopically and tumors were grown for 3 weeks. Tumor volume was measured after completion of the study. Knockdown of CCL2 in LLC cells resulted in a decrease in tumor volume when compared to non-targeting controls (Figure 4.1).
21 CCL2 knockdown alters immune cells of the TME
In order to determine the effects of CCL2 knockdown on cells of the tumor
microenvironment, mice were injected with LLC-NT or LLC-sh70 cells and tumors grew for 3 weeks. Flow cytometry using multicolor panels for identifying both T cells and monocytic cells were used to identify changes in population percentage. CCL2 knockdown in LLC cells resulted in an overall increase in recruited T cells, which can be broken down further into CD4+ and CD8+ populations (Figure 4.2).
Flow cytometry was also performed using a multicolor panel of markers for myeloid- derived cells (monocytes, macrophages, neutrophils). The MacB population is defined as CD11b+ cells that are then separated into distinct populations based on expression of CD64 and CD11c. MacB1 are classified lung monocytes, MacB2 are recruited monocytes, and MacB3 are recruited macrophages30. Knockdown of CCL2 in LLC cells resulted in a decrease in MacB2s and an increase MacB3s (Figure 4.3).
CCL2 knockdown in LLC cells confers sensitivity to anti-PD-1 therapy Next, to determine whether these effects have an impact on the efficacy of
immunotherapy, mice with tumors containing non-targeting or knockdown shRNA were treated with IgG2a controls or anti-PD-1 therapy. Tumors with CCL2 knockdown responded best to anti-PD-1 treatment, as exhibited by a decrease in tumor volume compared to controls. One interesting observation is the decreased tumor volume of LLC-NT tumors treated with anti-PD-1.
Traditionally, LLC cells do not respond well to immunotherapy, and it was expected that these tumors would be comparable in size to LLC-NT tumors treated with IgG2a. This difference in size may be due to poor orthotopic injection of the cells into the mouse lung, or an issue of the cells growing in culture before injection.
22
Overexpression of CCL2 in CMT167 cells has no impact on tumor volume
CMT167 cells, unlike LLC cells, express little to no CCL2 at baseline. Thus, we looked to determine if overexpression of CCL2 in these cells would give similar results seen to parental or non-targeting LLC cells. CMT167 cells were stably transfected with either an EGFP or CCL2 expression clones to generate overexpressing cell lines, which were then orthotopically injected into mice. CCL2 overexpression in CMT167 cells did not have significant on primary tumor volume (Figure 4.5). This may be due to the hostility of these cells, which form small primary tumors and metastasize quickly to the chest wall in the mouse, resulting in premature death.
Additional replicates must be repeated to determine whether overexpression of CCL2 would increase tumor volume.
Discussion and Future Directions
These experiments were conducted to test the hypothesis that knockdown of myeloid cell recruiting CCL2 would result in enhanced response to immunotherapy in an orthotopic mouse model; additionally, overexpression of CCL2 would exhibit the opposite effects. In LLC cells with CCL2 knockdown grew into smaller tumors than non-targeting controls. When tumors were treated with anti-PD-1 therapy, CCL2 knockdown tumors were more responsive than controls.
Enhanced T cell recruitment was also observed in cells containing CCL2 knockdown, and MacB populations were altered. Finally, CCL2 overexpression in CMT167 cells resulted in increased tumor burden compared to controls.
These studies used LLC cells where CCL2 was knocked down, indicating the importance of cancer-cell intrinsic expression and secretion of the chemokine. However, cancer cells are not the only cells that secrete CCL2 into the microenvironment, making it difficult to parse out the true effects of its altered expression. The next steps would be to generate a knockout mouse for
23
CCL2, eliminating its secretion from macrophages; tumors could then be grown in these mice and be treated with anti-PD-1 therapy. From this, we would learn how recruitment of monocytes and macrophages to the lung via CCL2 secretion affects the efficacy of immunotherapy.
In CCL2 knockdown tumors, T cell levels increased compared to controls. This suggests an alteration in expression and secretion of recruiting chemokines in the TME to attract more T cells. These would need to be identified and quantified; most likely they are CXCL9 and/or CXCL10. Tumor cells could be harvested and subjected to RT-PCR to identify changes in mRNA expression. Immunohistochemistry may also be performed using tissues harvested from non-targeting controls and CCL2 knockdown cells to measure chemokine levels.
MacB levels were altered in CCL2 knockdown cells compared to controls. This is most likely due to the elimination of the recruiting chemokine from the cancer cells. We observed a decrease in recruited immature monocytes, which are hypothesized to contribute to
immunotherapy resistance. Repetition of these experiments is required to determine whether the effects observed reach statistical significance.
24
Figure 4.1: Knockdown of CCL2 in LLC cells results in decreased tumor volume. LLC cells were stably transfected with non-targeting shRNA (LLC-NT) or shRNA targeting CCL2 (LLC- sh70) and were injected into wild type Black6 mice. Tumors grew for 3 weeks and lungs were removed to measure tumor volume in millimeters cubed (mm3). N=1.
Figure 4.2: CCL2 knockdown results in an increase in recruited T cells. LLC-NT and LLC- sh70 cells were injected into mice and tumors grew for 3 weeks. Tumor-bearing lungs were harvested, dissociated into a single-cell suspension and stained with a panel of T cell markers for flow cytometry. Values are percent total of all live cells. N=1.
25
Figure 4.3: CCL2 knockdown results in altered monocyte/macrophage recruitment to the TME. LLC-NT and LLC-sh70 cells were injected into mice and tumors grew for 3 weeks.
Tumor-bearing lungs were harvested, dissociated into a single-cell suspension and stained with a panel of myeloid cell markers for flow cytometry. MacB total values are a percentage of gated Ly6G- cells, and MacB1, 2, and 3 levels are the percentage of total MacBs. N=1.
26
Figure 4.4: Knockdown of CCL2 in LLC cells confers sensitivity to anti-PD-1 therapy.
LLC-NT and LLC-sh70 cells were injected into mice and established tumors for 7 days. Control IgG2a or anti-PD-1 antibody therapies were given IP for 2 weeks (4 doses total) at 8-10 mg/kg.
After 3 weeks, mice were sacrificed and primary tumor volume was measured. N=1.
Figure 4.5: Overexpression of CCL2 in CMT167 cells has little effect on tumor volume.
CMT167 cells containing EGFP control or CCL2 overexpression plasmid were injected into mice and tumors grew for 2 weeks. Tumor-bearing lungs were harvested, and tumor volume was measured (mm3). N=1.
27
Table 1: List of flow cytometry antibodies used in Chapter IV.
Flow Cytometry Antibodies Anti-mouse FcBlock eBioscience 14-0161-86
RRID AB_467135
T cell phenotypic, myeloid cell phenotypic Ki67-FITC eBioscience 11-5698-82
RRID AB_11151330
T cell phenotypic PD-1-PE eBioscience 12-9981-81
RRID AB_466289
T cell phenotypic IA/IE-PE/Dazzle594 eBioscience 12-5321-82
RRID AB_465928
T cell phenotypic, myeloid cell phenotypic CD3e-PerCP-Cy5.5 eBioscience 45-0031-82
RRID AB_1107000
T cell phenotypic CD69-PECy7 eBioscience 25-0691-81
RRID AB_469636
T cell phenotypic FoxP3-eF660 eBioscience 50-5773-82
RRID AB_11218868
T cell phenotypic CD45-AF700 eBioscience 56-0451-82
RRID AB_891454
T cell phenotypic, myeloid cell phenotypic CD8a-APC EF780 eBioscience 47-0081-82
RRID AB_1272185
T cell phenotypic CD4-eF450 eBioscience 48-0042-82
RRID AB_1272194
T cell phenotypic CD64-PE BD Biosciences 558455
RRID AB_647241
Myeloid cell phenotypic Cd11b-PE-Cy5 eBioscience 15-0112-82
RRID AB_468714
Myeloid cell phenotypic Ly6G-PE-Cy7 BD Biosciences 553310
RRID AB_1727561
Myeloid cell phenotypic SiglecF-A647 BD Biosciences 562680
RRID AB_2687570
Myeloid cell phenotypic CD11c-APC-Cy7 BD Biosciences 561241
RRID AB_10611727
Myeloid cell phenotypic Aqua Viability Dye ThermoFisher L34957 T cell phenotypic, myeloid
cell phenotypic
28
SUMMARY AND FUTURE DIRECTIONS
Lung cancer is the leading cause of cancer deaths in the United States, with 85% of cases being NSCLC. The poor 5-year survival rate of this disease has highlighted the need for new therapeutic combinations to be investigated. Cancer-cell specific MHCII has been implicated as an important marker for response to immunotherapies. Thus, induction of expression of surface MHCII on cancer cells in the TME may enhance response to these therapeutics. The studies in the thesis looked to determine how altering cancer-cell intrinsic properties, such as MHCII expression of CCL2 secretion, could alter the TME and promote tumor cell killing and reduced tumor burden.
Cancer cell-specific MHC Class II expression on cancer cells has been identified as a marker for enhanced response to immunotherapy, and for increased T cell recruitment. The in vitro experiments performed using EML4/ALK and KRAS mutant NSCLC cells have
demonstrated that MHCII expression can be induced via drug treatment. Use of the pan-HADC inhibitor TSA, MEK 1/2 inhibitor trametinib, or KRAS G12C specific AMG510, resulted in increases in CIITA expression. This transcriptional master regulator of MHCII expression acts as a readout as the message level. Flow cytometry experiments measuring cell surface expression of MHCII would be the next step to verify the RT-PCR data from Chapters 2 and 3. Each cell line responded differently to the combinations of drug, indicating that patients may experience differential response as well. In addition to their driving mutations, some cell lines have mutated p53, which can contribute further to these effects.
Knockdown of macrophage-recruiting chemokine CCL2 also showcased the effects of altering cancer cell-specific expression of genes. In LLC-sh70 cells containing the knockdown, tumor volume was greatly decreased, and tumors responded better to anti-PD-1 therapies than
29
controls. This is of particular interest, since LLC cells typically do not respond to
immunotherapy. Knockdown of CCL2 also led to an increased in T cells recruited to the TME, and altered recruited monocyte and macrophage populations. These results are promising, and they may be successful in enhancing response to immunotherapies in patients. There are clinically available CCR2 inhibitors that target and block the cognate receptor of CCL2. One example of these is CCX872, which was shown to be effective in combination with anti-PD-1 therapy in sensitizing glioblastoma to treatment. Tumor-associated MDSCs decreased, and median and overall survival improved in their mouse models31. With these results and support from research in other cancers, CCL2 inhibition may be pursued as a combination therapy with immune checkpoint inhibitors in NSCLC.
In the near future, we will need new strategies for treating NSCLC patients to improve the overall quality of life and survival rate. Novel combinations of therapeutics using clinically available drugs are the ideal route, and the use of HDAC inhibitors, MAP kinase inhibitors, and immunotherapy may offer success. Due to the heterogenous nature of the disease, it will be important to determine how oncogenic drivers impact response. There are a number of oncogenic drivers in NSCLC (KRAS, EML4/ALK, EGFR, ROS, MET, etc.) that need to be investigated in the future to understand the full scope of these treatments.
30
REFERENCES
1. Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2020. CA Cancer J Clin. 2020;70:7-30.
2. Ettinger DS, Beplar G, Bueno R, Chang A, Chang JY, et al. Non-small Cell Lung Cancer Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2006;4(6):548-82.
3. Borghaei H, Paz-Ares L, Horn L, Spigel DR. Nivolumab versus docetaxel in advanced nonsquamous non-small cell lung cancer. N Engl J Med. 2015;373:1627-1639.
4. Brahmer J, Reckamp KL, Baas P, Crinò L. Nivolumab versus docetaxel in advanced squamous-cell non-small cell lung cancer. N Engl J Med. 2015;373:123-135.
5. Garon EB, Rizvi NA, Hui R, Leighl N, et al. Pembrolizumab for the treatment of non- small cell lung cancer. N Engl J Med. 2015;372:2018-2028.
6. Seto E, Yoshida M. Erasers of Histone Acetylation: The Histone Deacetylase Enzymes.
Cold Spring Harb Perspect Biol. 2014;6:a018713.
7. Suraweera A, O’Byrne KJ, Richard DJ. Combination Therapy with Histone Deacetylase Inhibitors (HDACi) for the Treatment of Cancer: Achieving the Full Therapeutics Potential of HDACi. Front Oncol. 2018;8:92.
8. McClure JJ, Li X, Chou CJ. Advances and Challenges of HDAC Inhibitors in Cancer Therapeutics. Adv Cancer Res. 2018;138:183-211.
9. Hanahan D, Weinberg RA. Hallmarks of Cancer: The Next Generation. Cell.
2011;144(5):646-674.
10. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;252-264.
11. Johnson DB, Estrada MV, Salgado R, Sanchez V, Doxie DB, et al. Melanoma-specific MHC-II expression represents a tumor-autonomous phenotype and predicts response to anti-PD-1/PD-L1 therapy. Nat Comm. 2016;7:10582.
12. Johnson AM, Bullock BL, Neuwelt AJ, Poczobutt JM, Kaspar RE, et al. Cancer Cell- Intrinsic Expression of MHC Class II Regulates the Immune Microenvironment and Response to Anti-PD-1 Therapy in Lung Adenocarcinoma. J Immunol. 2020;204:2295- 307.
13. Axelrod ML, Cook RS, Johnson DB, Balko JM. Biological Consequences of MHC-II Expression by Tumor Cells in Cancer. Clin Cancer Res. 2019;25:2392-402.
14. ClinicalTrials.gov. U.S. National Library of Medicine. Phase II Anti-PD1 Epigenetic Therapy Study in NSCLC. (NA_00084192). 2013; Identifier NCT01928576.
31
15. ClinicalTrials.gov U.S. National Library of Medicine. A Phase 1b Dose
Escalation/Expansion Study of Abexinostat in Combination with Pembrolizumab in Patients with Advanced Solid Tumor Malignancies. 2018; Identifier NCT03590054.
16. ClinicalTrials.gov U.S. National Library of Medicine. Phase 2 Study of Glesatinib, Sitravatinib or Mocetinostat in Combination with Nivolumab in Non-Small Cell Lung Cancer. 2016; Identifier NCT02954991.
17. ClinicalTrials.gov U.S. National Library of Medicine. Pembro and Vorinostat for Patients with Stage IV Non-Small Cell Lung Cancer (NSCLC). 2015; Identifier NCT02638090.
18. Booth L, Roberts JL, Poklepovic A, Kirkwood J, Dent P. HDAC inhibitors enhance the immunotherapy response of melanoma cells. Oncotarget. 2017;8(47):83155-83170.
19. Zheng H, Zhao W, Yan C, Watson CC, Massengill MM, et al. HDAC Inhibitors Enhance T-Cell Chemokine Expression and Augment Response to PD-1 Immunotherapy in Lung Adenocarcinoma. Clin Cancer Res. 2016;22(16):4119-32.
20. Tay RE, Olawoyin O, Cejas P, Xie Y, Meyer CA, et al. Hdac3 is an epigenetic inhibitor of the cytotoxicity program in CD8 T Cells. J Exp Med. 2020;217(7):e20191453.
21. Idso JM, Lao S, Schloemer NJ, Knipstein J, Burns R, et al. Entinostat augments NK cell functions via epigenetic upregulation of IFIT1-STING-STAT4 pathway. Oncotarget.
2020;11(20):1799-1815.
22. Roumenina LT, Daugan MV, Petitprez F, Sautes-Fridman C, Herman Fridman WH.
Context-dependent roles of complement in cancer. Nat Rev Cancer. 2019;19(12):698- 715.
23. Kleczko EK, Kwak JW, Schenk EL, Nemenoff RA. Targeting the Complement Pathway as a Therapeutic Strategy in Lung Cancer. Front. Immunol. 2019;10:954.
24. Reis ES, Mastellos DC, Ricklin D, Mantovani A, Lambris JD. Complement in cancer:
untangling an intricate relationship. Nat Rev Immunol. 2018;18(1):5-18.
25. Neuwelt AJ, Kimball AK, Johnson AM, Arnold BW, Bullock BL, et al. Cancer-cell intrinsic expression of MHC II in lung cancer cell lines is actively restricted by MEK/ERK signaling and epigenetic mechanism. J Immunother Cancer.
2020;8(1):e000441.
26. Maddalo D, Manchado E, Concepcion CP, Bonetti C, Vidigal JA, et al. In vivo
engineering of oncogenic chromosomal rearrangements with the CRISPR/Cas9 system.
Nature. 2014;516(7531):423-427.
32
27. Molina-Arcas M, Moore C, Rana S, van Maldegem F, Murarza E, et al. Development of combination therapies to maximize the impact of KRAS-G12C inhibitors in lung cancer.
Sci Transl Med. 2019;11(510):eaaw7999.
28. Bullock BL, Kimball AK, Poczobutt JM, Neuwelt AJ, Li HY, et al. Tumor-intrinsic response to IFNγ shapes the tumor microenvironment and anti-PD-1 response in NSCLC.
Life Sci Alliance. 2019;2(3):e201900328.
29. Li HY, McSharry M, Bullock B, Nguyen TT, Kwak J, et al. The Tumor
Microenvironment Regulates Sensitivity of Murine Lung Tumors to PD-1/PD-L1 Antibody Blockade. Cancer Immunol Res. 2017;5(9):767-777.
30. Poczobutt JM, De S, Yadav VK, Nguyen TT, Li H, et al. Expression Profiling of Macrophages Reveals Multiple Populations with Distinct Biological Roles in an
Immunocompetent Orthotropic Model of Lung Cancer. J Immunol. 2016;196(6):2847-59.
31. Flores-Toro JA, Luo D, Gopinath A, Sarkisian MR, Campbell JJ, et al. CCR2 inhibition reduces tumor myeloid cells and unmasks a checkpoint inhibitor effect to slow
progression of resistant murine gliomas. Proc Natl Acad Sci U S A. 2020;117(2):1129- 1138.
