The Role of Natural Killer Cells in the Context of Oncolytic Herpes Simplex Virotherapy

for Glioblastoma

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By

Christopher Allen Alvarez-Breckenridge

Graduate Program in Integrated Biomedical Science Program

The Ohio State University

2011

Committee:

Dr. E. Antonio Chiocca MD, PhD “Co-Advisor”

Dr. Michael Caligiuri, MD “Co-Advisor”

Dr. Balveen Kaur, PhD

Dr. Susheela Tridandapani, PhD

Copyright by

Christopher Allen Alvarez-Breckenridge

2011

Abstract

It is controversial as to whether the host immune response hinders or improves the efficacy of oncolytic Herpes Simplex viral (oHSV) therapy of glioblastoma (GBM).

Natural killer cells (NK) limit viral infections, and previous work suggests they may similarly attenuate virotherapy. Using both xenograft and syngeneic intracranial GBM tumor models, we used flow cytometry to evaluate the temporal pattern and phenotypic characteristics of NK cells present in the periphery and recruited to the site of oHSV infection. Within hours after infection and continuing through 72 hours following oHSV inoculation, NK cells were rapidly recruited to tumor bearing hemispheres. Moreover, these NK cells exhibited an activated phenotype, including enhanced CD69, CD62L,

CD27, NKG2D, and Ly49D staining compared to vehicle treated mice. However, neither the number nor phenotype of peripheral NK cells was altered following oHSV infection.

This robust NK response was confirmed to be detrimental to OV efficacy through the enhanced survival of NK depleted mice inoculated with oHSV compared to oHSV treated mice possessing NK cells. Interestingly, oHSV treated mice exhibited robust macrophage recruitment and activation at the site of infection. This was accompanied by the induction of macrophage/microglial associated inflammatory gene and protein expression, including CXCL9, CXCL10, CXCL11, iNOS, and TNF-α. However, when

ii mice were depleted of their NK cells or IFN-γ knockout mice were used, their expression was abrogated.

In vitro, human NK cells preferentially lysed oHSV-infected GBM in a cell contact, perforin, and DNAM-1 dependent manner. Fusion proteins were used to detect currently unknown ligands for the NK natural cytotoxicity receptors (NCR) and decipher the critical NK activating ligands that mediate this response. Following oHSV infection of a panel of GBM stem cells and cell lines, we detected robust up-regulation of ligands for

NKp46 and NKp30. GFP expression was used to discriminate oHSV infected GBM cells and preferential NKp30/NKp46 ligand expression was found in the GFP+ population of cells. Additionally, blocking antibodies against either NKp30 or NKp46 abrogated NK mediated clearance of oHSV infected GBM, while antibodies against NKp44 did not inhibit killing.

We have previously shown that immunomodulation with cyclophosphamide (CPA) and valproic acid (VPA) enhances oHSV efficacy. CPA administered prior to virus inoculation abrogated the oHSV induced NK and macrophage recruitment into the tumor at all time points tested compared to oHSV alone. Similarly, VPA treatment resulted in a decline in NK and macrophage recruitment at 6 and 24 hours post-oHSV; however, a robust increase at 72 hours-post-oHSV was seen, resembling the response seen with oHSV alone. VPA was also found to have a profound immunosuppressive effect on human NK cells in vitro. NK cytotoxicity was abrogated following exposure to VPA

iii through down-modulation of cytotoxic gene expression and granzyme B protein levels.

In addition, IFN-γ was suppressed in a Stat5/T-bet dependent manner.

Collectively, these findings demonstrate that oHSV therapy for GBM is limited in part by a robust NK cell response mediated by specific NCRs, uncovering novel potential targets to enhance cancer virotherapy. Moreover, pharmacological co-therapies, such as CPA and VPA with oHSV, alter the host immune response to the virus albeit in differing ways.

Future work will be needed to further define the nature of the innate immune response, how it coordinates downstream anti-tumor immunity, and how pharmacological agents can be optimized to modulate the host response to oHSV.

iv

Dedication

This document is dedicated to those who have stood by my side on this long journey. In particular, I would like to thank my undergraduate advisor, Dr. Charis Eng, for her planting the idea of an MD/PhD; Dr. Gregory Jusdanis for the encouragement and support he has given me for so many years and ultimately on this quest to γνῶθι σεαυτόν; and to Dr. Lauren Walters—thank you for sticking with me!

I would certainly not be here without the devotion of my parents, Carmen Alvarez-

Breckenridge and David Breckenridge. This dissertation is a testament to their love, prayers, and support. I would like to thank my wife, Jennifer Alvarez-Breckenridge, for her love, unceasing support of my academic pursuits, and for always showing me that the glass is half full.

Lastly, I would never have embarked on this road had I not experienced the call to medicine during my time in Lourdes. I dedicate this work and my future endeavors to

Our Lady of Lourdes, and her son, Jesus Christ.

v

Acknowledgments

I would like to thank my mentors Dr. E. Antonio Chiocca and Dr. Michael Caligiuri for all of their help and support during my PhD. In particular, I would like to thank them for the opportunity to pursue this project; for constantly challenging me on every step along the way in order to achieve my goal of becoming an independent scientist; and for empowering me to exceed beyond my self-constructed limits.

I would also like to thank Dr. Jianhua Yu for being by my side throughout this journey and never showing the slightest bit of frustration despite the numerous questions I have asked him over the years. Without your help, I could have never taken this project off the ground—thank you!

The entire Dardinger Laboratory, especially: Dr. Balveen Kaur, Dr. Sean Lawler, Rick

Price, Kazue Kasai, Dr. Hiroshi Nakashima, Dr. Kazuo Okemoto, Amy Hasely, Jason

Pradarelli, Jayson Hardcastle, Dr. Nina Dmitrieva, Dr. Kazuhiko Kurozumi. In addition, a special thank you to the Dardinger support staff, in particular Melinda Akins, Erica

Chambers, and Louvenia Broadnax. To the entire Caligiuri Laboratory, especially: Susan

McClory, Min Wei, Charlene Mao, Dr. Tom Liu, Dr. Rossana Trotta, Tiffany Hughes,

Nick Zorko, Edward Briercheck, and Dr. Jason Chandler.

vi

To the Integrated Biomedical Graduate Program and the Medical Scientist Program for their support and guidance, particularly Dr. Alan Yates, Dr. Larry Schlesinger, Dr.

Lawrence Kirschner, and Ashley Bertran.

I would like also like to thank my collaborators for their help in this project: Dr. Soledad

Fernandez, Dr. Susheela Tridandapani, Dr. Alessandro Moretta, Dr. Ofer Mandelboim, and Dr. Eric Vivier.

I would also like to thank my committee members for constantly challenging me and fostering my scientific acumen.

Lastly, I thank the NIH, AMA Foundation, and Ohio State’s Center for Clinical and

Translational Science for their financial support.

vii

Vita

2001...... St. Charles Preparatory High School

2005...... B.S. Biology, The Ohio State University

2005...... B.A. Classics, The Ohio State University

2005-present...... Graduate Research Associate, Department

of Neurological Surgery, The Ohio State

University

Publications

1. Alvarez-Breckenridge, C., Waite K.A., Eng, C. (2007) PTEN Regulates

Phospholipase D and Phospholipase C. Hum. Mol. Genet., 16, 1157-1163

2. Haseley, A., Alvarez-Breckenridge, C., Chaudhury A.R., Kaur, B. (2009)

Advances in oncolytic virus therapy for glioma. Recent Pat. CNS Drug Disc., 4

(1): 1-13

3. Alvarez-Breckenridge, C., Kaur, B., Chiocca, E.A. (2009) Pharmacologic and

chemical adjuvants in tumor virotherapy. Chem Rev., 109 (7): 3125-40

viii

4. Yu, J., Mitsui, T., Wei, M., Mao, H., Butchar, J.P., Shah, M.V., Zang, J., Mishra,

A., Alvarez-Breckenridge, C., Liu, X., Liu, S., Yokohama, A., Trotta, R.,

Marcucci, G., Benson, D.M., Loughran, T.P., Tridandapani, S., Caligiuri, M.A.

(2011) NKp46 identifies an NKT cell subset susceptible to leukemic

transformation in mouse and human. J Clin Invest., Epud ahead of print.

Field of Study

Major Field: Integrated Biomedical Science Program

ix

Table of Contents

Abstract...... ii

Dedication...... v

Acknowledgments...... vi

Vita...... viii

List of Tables ...... xiv

List of Figures...... xv

Chapter 1: Introduction...... 1

Section 1—Current treatment approaches for malignant glioma...... 1

Section 2—OV clinical trials for glioma...... 3

Section 3—The host response to OV therapy ...... 6

Section 4—NK response to virus and tumors ...... 8

Section 5—Pharmacologic modulation of the host response to OV...... 11

Chapter 2: Background ...... 18

Section 1—Glioblastoma overview ...... 18

Section 2—Glioblastoma and the immune environment ...... 20 x

Section 3—Oncolytic viruses as a therapeutic modality...... 21

Section 4—The immediate host response following OV infection...... 24

Section 4.1—Barriers to viral entry...... 26

Section 4.2—The intracellular antiviral response ...... 28

Section 4.3—Extracellular tumor microenvironment barriers ...... 31

Section 4.4—Immune responses to oncolytic viruses and its cellular mediators...... 32

Section 5—Immune response to HSV infection ...... 36

Section 6—Intracellular pathways affecting virotherapy...... 38

Section 7—Pharmacological modulation of host factors to enhance OV therapy...... 39

Section 7.1—Immune modulators...... 40

Section 7.2—Histone deacetylase inhibitors...... 43

Section 7.3—Antiangiogenic agents ...... 45

Section 8—Placing NK cell biology in the OV context...... 47

Section 9—NK deficiencies...... 50

Chapter 3: The in vivo NK cell response to oHSV is detrimental to viral oncolysis...... 58

Introduction ...... 58

Methods...... 60

Results ...... 65

xi

Oncolytic virus induces rapid natural killer cell recruitment in both human xenograft

and syngeneic murine glioblastomas...... 65

oHSV inoculation induces an activated NK phenotype...... 66

oHSV therapy induces the recruitment of distinct cytotoxic NK subsets...... 67

Macrophages and microglia are activated following oHSV therapy in a NK

dependent manner...... 68

Depletion of NK cells in vivo leads to enhanced viral replication and improves

glioblastoma therapy...... 70

Discussion ...... 72

Table 7: continued...... 119

Chapter 4: NK cells kill oHSV infected glioblastoma cells using NKp30 and NKp46 . 121

Introduction ...... 121

Methods...... 123

Results ...... 126

Human glioblastoma activate NK cells from human donors...... 126

NK cells preferentially kill oHSV infected glioblastoma in both mouse and human

models...... 127

NK mediated killing of oHSV infected glioblastoma is dependent upon cell contact,

perforin, and DNAM-1...... 128

xii

The NCRs NKp30 and NKp46 mediate clearance of oHSV-infected glioblastoma.

...... 129

Discussion ...... 131

Chapter 5: The innate immune response is attenuated by valproic acid co-administration with oHSV ...... 155

Introduction ...... 155

Methods...... 157

Results ...... 161

Valproic acid suppresses immune cell infiltration following oHSV administration161

oHSV inoculation with VPA reduces the inflammatory response ...... 162

VPA suppresses IFN-γ production in a Stat5/T-bet dependent fashion ...... 163

Mediators of NK cell cytotoxicity are reduced by VPA ...... 164

NK cell mediated killing of oHSV infected glioblastoma is suppressed by VPA .. 164

Discussion ...... 165

Chapter 6: Future directions...... 182

References...... 188

xiii

List of Tables

Table 1.1: Clinical trials using oncolytiv viruses for the treatment of malignant gliomas.

...... 16

Table 2.1: Features of oncolytic viruses being used for glioma therapy ...... 57

Table 3.1: Phenotype of NK cells recruited to the site of infection in glioma xenograft106

Table 3.2: Phenotype of NK cells recruited to the site of infection in syngeneic glioma model...... 107

Table 3.3: Developmental status of recruited NK cells following oHSV infection ...... 108

Table 3.4: Changes in inflammatory cytokine profile between vehicle and rQnestin34.5 treated mice...... 109

Table 3.5: Changes in inflammatory cytokine profile between rQnestin4.5 and

Asialo+rQnestin34.5 treated mice ...... 115

xiv

List of Figures

Figure 1.1: Human NK cell receptors and ligands...... 15

Figure 2.1: Understanding and targeting the host response to oncolytic viral therapy is needed for its clinical success...... 53

Figure 2.2: Viral infection elicits a variety of antiviral cellular responses...... 54

Figure 2.3: The tumor microenvironment following oncolytic viral infection undergoes dynamic changes that can be targeted with pharmacological co-therapy...... 55

Figure 2.4: Viral infection induces an NK mediated inflammatory response...... 56

Figure 3.1: Construction of an oHSV that replicates specifically in tumor cells expressing nestin...... 77

Figure 3.2: NK cells recruitment to the site of oHSV infection increases in a time dependent manner...... 78

Figure 3.3: The infiltrative NK response occurs in both xenograft and syngeneic glioblastoma models...... 80

Figure 3.4: The NK response to oHSV is recapitulated in a syngeneic model...... 81

Figure 3.5: Oncolytic and wild-type viral infection results in a similar temporal NK response...... 82

Figure 3.6: Active viral replication is required for NK cell recruitment...... 83

xv

Figure 3.7: NK cells are activated following oHSV therapy...... 84

Figure 3.8: Peripheral NK cells do not exhibit phenotypic changes following oHSV therapy...... 85

Figure 3.9: oHSV administration results in the recruitment of distinct NK subsets...... 86

Figure 3.10: Viral infection and glioblastoma formation result in the recruitment of differential NK subsets...... 88

Figure 3.11: Macrophages and lymphocytes recruited to the site of oHSV infection increases in a time dependent manner...... 89

Figure 3.12: Macrophages are activated following oHSV therapy...... 90

Figure 3.13: NK depletion does not alter the proportion of macrophages following oHSV administration...... 91

Figure 3.14: NK cells mediate macrophage and microglia activation following oHSV inoculation...... 92

Figure 3.15: NK cells mediate TNF-α production from microglia and macrophages. .... 94

Figure 3.16: iNOS production following oHSV infection is induced in an IFN-γ dependent manner...... 95

Figure 3.17: NK cells and IFN-γ mediate CXCL9, 10, 11 expression...... 97

Figure 3.18: NK depletion does not alter microglia or macrophage production of IFN-γ inducible chemokine...... 99

Figure 3.19: Asialo-GM1 treatment depletes NK cells...... 100

Figure 3.20: NK depletion enhances oHSV efficacy...... 102

xvi

Figure 3.21: NK cells mediate robust inflammatory response following oHSV infection.

...... 104

Figure 3.22: A lack of HSV infectability in C57BL/6 mice prevents assessment of oHSV efficacy in the KR158dEGFR glioblastoma model...... 105

Figure 4.1: NK cells are activated following oncolytic viral infection of co-cultured target cells...... 136

Figure 4.2: IL-15 is expressed following oncolytic viral infection...... 139

Figure 4.3: NK cells preferentially kill glioblastoma cells infected with oncolytic virus.

...... 140

Figure 4.4: NK cells preferentially kill virally infected glioblastoma cells at early timepoints after infection...... 142

Figure 4.5: Cytotoxic T lymphocytes and NK cells exhibit a differential cytotoxic profile against virally infected glioblastoma cells...... 143

Figure 4.6: Exogenous IL-15 and TGF-β inversely regulate NKp30 expression...... 144

Figure 4.7: NK killing oHSV infected glioblastoma requires cell contact but occurs independently of NKG2D or MHC I modulation...... 145

Figure 4.8: NK cell killing of virally infected cells occurs primarily through a perforin mediated mechanism...... 146

Figure 4.9: Ligands for NK cells are endogenously expressed on glioblastoma...... 147

Figure 4.10: Human NK cells clear oHSV infected glioblastoma through NKp30 and

NKp46...... 149

xvii

Figure 4.11: Murine NK cells kill virally infected glioblastoma cells in an NKp46 dependent manner...... 150

Figure 4.12: OV infection of glioblastoma induces novel ligands for NKp30 and NKp46.

...... 151

Figure 4.13: Induction of NCR expression occurs after oHSV infection and TMZ exposure but not other forms of cellular stress ...... 153

Figure 5.1: VPA attenuates NK recruitment initially after oHSV infection...... 169

Figure 5.2: VPA attenuates macrophage recruitment initially after oHSV infection..... 171

Figure 5.3: VPA suppresses NK activation initially after oHSV infection...... 172

Figure 5.4: VPA attenuates IFN-γ expression following oncolytic viral infection...... 173

Figure 5.5: VPA reduces human NK cell expression of IFN-γ transcript...... 174

Figure 5.6: VPA attenuates cytokine mediated induction of NK cell IFN-γ secretion... 175

Figure 5.7: VPA antagonizes cytokine induced Stat5 and Tbet upregulation...... 177

Figure 5.8: VPA suppresses NK cell cytotoxicity genes in a dose dependent manner. . 178

Figure 5.9: VPA reduces granzyme B protein levels in a dose dependent manner...... 179

Figure 5.10: NK cell cytotoxicity is suppressed following exposure to VPA...... 180

xviii

Chapter 1: Introduction

Section 1—Current treatment approaches for malignant glioma

Despite active research in the field of glioblastoma (GBM), the current prognosis for patients diagnosed with this disease is particularly poor, with only anecdotal examples of

“response”1,2. The first line of treatment for glioblastoma consist of initial surgical removal of the tumor mass. Advances in intraoperative imaging modalities have facilitated significant tumor debulking; however, the histologic nature of GBM makes full tumor removal impossible. The implantation of Gliadel wafers, consisting of the chemotherapeutic carmustine, during surgery is FDA approved and has provided a modest increase in overall survival3. Following surgery, radiation is administered to the resected cavity and surrounding brain in order to eradicate infiltrative tumor cells that were not visible during the initial tumor resection. In conjunction with standard therapy of 60 Gy radiation, the chemotherapeutic temozolomide is orally administered and has been shown to improve overall survival, particularly in patients with MGMT promoter methylation4,5.

Taken together, the combination of radiation and a chemotherapy regimen has extended patient survival form 3-4 months with surgery alone to 12-15 months with surgery, radiation, and chemotherapy3,4,6. Despite this extension in patient survival,

1 approximately 90% of patients experience a recurrence at the original tumor site7. Upon recurrence, optimal therapy is less defined. As a result, both the modest therapeutic efficacy of current standard of care for glioblastoma and an undefined approach for treating recurrent tumor highlights the importance of developing new therapies for this devastating disease.

A variety of novel radiation, surgical, chemical, immunological, and biological based approaches are currently being investigated to treat glioblastoma. For instance, by elucidating the signals responsible for tumor formation and pathogenesis, investigators are attempting to tailor inhibitors of aberrant cell signaling pathways that are present in glioblastoma. Despite the over-expression of EGFR, PDGFR, and VEGFR in many

GBMs, the redundancy of signaling pathways that are operating in these cells make a single therapeutic agent approach difficult to realize8. Alternatively, the highly vascular nature of glioblastoma makes this form of cancer a logical target for antiangiogenic treatment. While these treatment approaches have varying degrees of potential, particularly promising families of biological agents that have been intensely investigated for nearly two decades consist of oncolytic viruses (OV).

These naturally occurring and biologically engineered viruses, which are designed to replicate exclusively within tumors and culminate in the destruction of the cancer cells9,10, are currently being pursued in a variety of preclinical rodent models of cancer11-

14. Examples of these mutant viruses include both RNA (VSV15-20, Measles21-23,

2

Reovirus24-26, Sindbis27,28, Poliovirus29-32, NDV33-35, Retrovirus36,37) and DNA (Herpes

Simplex Virus38-48, Adenovirus49-60, Vaccinia61-65, Myxoma66,67) viruses. These initial studies have demonstrated strong evidence for the efficacy, safety, and tolerability of this approach, warranting thorough testing in early phase clinical trials68-76. Notably, China has recently approved the first OV for human use68.

One strain of virus that has received significant attention as an oncolytic virus is herpes simplex virus type 1 (HSV-1). HSV-1 based oncolytic viruses (oHSV) are a particularly encouraging therapeutic modality for several reasons. They posses: 1) a wide range of infectability for various tumors; 2) an inherent cytolytic nature; 3) a well characterized genome with many non-essential genes that can be manipulated with the insertion of novel therapeutic genes; and 4) susceptibility to widely available anti-herpetic drugs which can limit the presence of undesirable replication77,78. Moreover, its neurotropism makes it an ideal choice for treating human nervous system tumors. Based on these properties and encouraging preclinical in vitro and in vivo results, oHSV therapy promises to be a viable therapeutic option that warrants examination in the clinical setting.

Section 2—OV clinical trials for glioma

Clinical safety has been demonstrated through several phase I trials for recurrent glioblastoma. Five different clinical trials have tested one of two different strains of attenuated HSV-1: G20779,80 and 171681-83. For each of these trials, a maximum tolerated

3 dose was neither achieved nor was there any evidence of toxicity that could be attributed to the administered virus. Additionally, oncolytic adenovirus69, Newcastle disease virus84, and reovirus85 have been shown to be safe in dose escalation trials (Table 1.1).

These findings indicate the relative safety of this approach in human brains. However, findings from these trials also point out several shortcomings to in vivo OV therapy that must be addressed in order translate this promising therapeutic modality into an effective therapy in humans.

The most recent clinical trial for oncolytic HSV in recurrent glioblastoma patients demonstrates several salient findings that raise important points for future preclinical investigation and subsequent clinical trials. In this trial, Markert et al. used G207, an oncolytic herpes virus, which is missing both copies of the neurovirulence γ34.5 loci and has a gene disrupting insertion of the beta galactosidase gene within the UL39 locus encoding the virus’ ribonucleotide reductase gene. This virus was inoculated into the brain before and after tumor resection. While a previous clinical trial demonstrated the safety of G20779, this subsequent study attempted to determine evidence of HSV replication and the extent of immune infiltration into the tumor microenvironment following OV administration80. While viral replication was observed in a portion of the patients enrolled in the study, there was a significant amount of variability in viral levels between patients. These findings are in contrast to the assumption in the field of OV therapy that even small initial inoculums should amplify significantly following successive rounds of viral replication86-89. Additionally, immunohistochemical (IHC)

4 analysis of tumor tissue demonstrated that G207 administration elicited a robust increase in CD3, CD8, CD20, and HAM56 staining. Notably, staining with the pan-NK maker,

CD56, was not included. These IHC findings are consistent with preclinical findings of our laboratory90,91 and suggest that OV infection/replication causes the recruitment of inflammatory infiltrates into the site of viral infection.

Additionally, testing of the genetically engineered mutant adenovirus (ONYX-015) in a phase I trial for glioblastoma provided similar results. While this treatment was confirmed to be safe and a maximum tolerated dose was not reached, an overall survival benefit was not observed92. Notably, when biopsies were taken from recurrent tumors, a significant inflammatory, mononuclear infiltrate was observed within the tumor microenvironment. The nature of this histological feature, similar to that seen by Markert et al., is uncertain. For example, it is unclear whether these findings correlate with a potential benefit by stimulating an anti-tumor immune response or a detriment by eliminating OV initially after infection thereby preventing initial rounds of viral replication within the tumor.

Taken together, the findings from these clinical trials demonstrate the need for a clearer understanding of the host-based factors and cellular mediators that are responsible for limiting viral infection, replication, and spread. By clarifying the host response to the virus, subsequent clinical trials can be designed to modulate these obstacles to viral propagation and achieve enhanced OV efficacy.

5

Section 3—The host response to OV therapy

Based on the observations of initial clinical trials, we hypothesize that the rapid inflammatory response that is elicited following OV infection is one of the critical barriers that prevents initial viral infection, replication, and survival within the injected tumor. By limiting virotherapy in this way, circumventing this inflammatory response has the potential of significantly enhancing initial viral replication and dissemination.

Preclinical in vivo findings have demonstrated that within a few days following OV inoculation, viral spread is attenuated and viral titers decrease as a function of time86,90.

In order to explain these findings, it is critical to understand how the oncolytic virus interacts in the context of the tumor microenvironment and identify the critical factors that could potentially limit viral replication. Barriers to effective viral propagation and tumor clearance include barriers to viral entry; intracellular signaling and antiviral defenses; the extracellular tumor environment; and the initial immune response to virus.

Following the initial detection of the OV by the host, a series of intracellular signaling cascades are activated. While there are a variety of cellular methods for detecting the foreign pathogen, an antiviral state is ultimately achieved through the production of type I interferons (IFN-I). Although IFN-I is not intrinsically antiviral, its production is responsible for the activation of interferon regulatory factors and interferon stimulatory genes that are integral for creating an antiviral state93-98. Moreover, experimental models

6 that have depleted IFN ligand resulted in robust viral amplification, demonstrating the importance of IFN-I as an antiviral mediator99.

In addition to the intracellular mechanisms for detecting OV, a classic response to wild- type viral infection is vasodilation and hyperpermeability100,101. This physiologic response integrates the tumor microenvironment with the inflammatory host response that is elicited shortly after viral infection. During the inflammatory response, peripheral leukocytes bind to the vascular lining and induce endothelial activation. The end result of this process is hyper-permeability of vasculature, increased tissue edema, and perivascular inflammatory cell infiltration. This vascular leakage has been both observed following OV treatment and confirmed to be detrimental to OV propagation 91.

The rapid response of innate immunity, consisting of natural killer cells (NK), monocytes, macrophages, and neutrophils, provides an initial, potent line of defense for the host and limits initial viral infection, replication, and spread; facilitates the maturation of antigen presenting cells; and communicates with the adaptive arm of the immune system to regulate its response. Due the rapid decline in viral titers within days of inoculating various oncolytic viruses17,90, the innate immune response has been implicated as a critical factor in this response. In fact, the clearance of over 80% of

HSV-derived OV particles has been found to correspond with the rapid recruitment of peripheral monocytes into the site of viral infection, suggesting that these cells participate in mediating OV clearance45,90,102. These findings were validated through a study that

7 examined the impact of macrophage depletion on OV efficacy86. In both this study and an additional set of work that investigated neutrophil depletion103, investigators have clearly demonstrated increased viral titers and enhanced tumor clearance in the absence of these inflammatory mediators. These findings implicate the cellular components of innate immunity as critical mediators of the initial antiviral response that limits OV efficacy.

Section 4—NK response to virus and tumors

While a variety of studies have elucidated the cellular contributions of the host response to OV, a detailed examination into the NK response to oncolytic HSV has been lacking.

Interestingly, profound human NK cell deficiencies have led to overwhelming herpes viral infections, supporting the notion that this innate immune effector cell has specific recognition of, and control over, this viral infection104-106. The relevancy of this cell population in our model was published from our laboratory’s finding that NK cells are rapidly recruited into the CNS following oHSV administration90; however, the nature of this viral-induced NK response is unclear. For example, what role do NK cells have in recruiting activated macrophages and microglia following oHSV therapy? Does oHSV used for the treatment of GBM induce a different NK activation profile compared to wild-type HSV infection? Does oHSV infection of GBM lead to the preferential NK- mediated clearance of these virally infected cells compared to uninfected GBM? Are there discrepancies between activated NK cells that are recruited in mice bearing xenograft GBM versus syngeneic GBM? Lastly, can we temporarily pharmacologically

8 modulate the NK immune response to oHSV-infected GBM cells in order to enhance oHSV therapeutic efficacy?

NK cells are able to carry out their diverse repertoire of activities through a detection system that relies on the engagement of a variety of cell surface activating and inhibitory receptors on NK cells that bind MHC class I and class I-like molecules (Figure 1.1).

Through the binding of these receptors, a dynamic equilibrium is achieved that differentiates the recognition of “self” cells from malignantly or virally transformed target cells. The role of NK cell mediated clearance of transformed GBM cells is currently unresolved. GBMs are readily killed by NK cells in vitro107 and despite an intense immunosuppressive tumor microenvironment as the GBM progresses, a recent report found that peripheral NK numbers were not altered in patients with GBM108. NK cells have also been demonstrated to preferentially lyse virally infected target cells through either the elaboration of cytotoxic granules containing lytic enzymes or through the binding of apoptosis-inducing receptors on target cells109. Through the expression of both virally and tumor/stress associated ligands for NK activating receptors, a target cell such as an infected GBM can become susceptible to NK mediated lysis. Countering this is the production of TGF-β and related molecules by the refractory GBM cells that impedes both NK cell production of IFN-γ110 and the ability of NK cells to directly lyse tumor cell targets111.

Recent discoveries have aimed to elucidate the role of HSV-1 infection in modulating

9

NK activating ligand expression. HSV-1 infection of fibroblast cells, through its immediate early gene product ICP0, resulted in increased susceptibility to NK mediated lysis in a MHC-I independent fashion112. NK activation was achieved independently of the NK activating receptor NKG2D112. Rather, NK stimulation was elicited through the expression of an unknown ligand(s) of the natural cytotoxicity receptors (NCR) on the cell surface of infected cells. Moreover, HSV infection was subsequently found to downmodulate NKG2D ligand expression due to late viral gene products113. These findings demonstrate the modulation of ligands for NK activating receptors following wild type HSV infection; however, neither the identity of these ligands nor the nature of their cooperative binding to NK activating receptors is currently resolved.

Beyond cell-mediated viral clearance, NK cells are potent producers of IFN-γ110. In the context of oHSV therapy, the detrimental role of this antiviral cytokine has been uncovered through an IFN-γ depletion study that resulted in enhanced intratumoral viral titers90. Additionally, IFN-γ is vitally important in the activation of macrophages, thereby facilitating their ability to kill both viral and obligate intracellular pathogens114.

Collectively, these findings demonstrate the critical nature of NK cells in coordinating the antiviral response to oHSV therapy.

A number of groups have also studied the ability of human glioblastomas to induce NK cell cytotoxicity. Based on their expression profile of ligands for activating NK receptors107,115, NK cells have been demonstrated to actively lyse a variety of

10 glioblastoma cells. Combining findings that NK cells are activated following HSV-1 infection and that NK ligands are endogenously expressed on cultured gliomas results in several questions in the context of oHSV therapy for glioblastomas. For instance, will oncolytic HSV infection of glioblastoma result in their preferential killing by NK cells?

This is a particularly relevant question since glioblastomas are endogenously potent activators of NK cells. By addressing this question, it will be possible to differentially assess the contributing signals that are unique to viral infection and subsequent NK activation.

By uncovering the unique NK ligand signature following oHSV infection and deciphering the receptor-ligand interactions that are responsible for the preferential clearance of virally infected cells over uninfected GBMs, it should be possible to delineate the critical signals responsible for the NK mediated anti-oHSV response. This understanding will pave the way for the subsequent development of therapeutic approaches to selectively target and block the NK receptors implicated in the anti-oHSV response. Conversely, if our hypothesis is incorrect and the anti-tumor NK response to oHSV therapy is therapeutically beneficial, the critical receptor—ligand interactions have the potential of being important therapeutic targets for downstream immunotherapy.

Section 5—Pharmacologic modulation of the host response to OV

Although the host immune response has been demonstrated to participate in limiting initial viral propagation following OV inoculation, there are a variety of therapeutic

11 approaches that can be used to attenuate this response and thereby synergize with OV therapy. For instance, the complement system is rapidly activated following viral infection. This response is intended to directly neutralize virus and clear virally infected cells. Targeting the complement system with various pharmacologic agents has proven useful towards enhancing OV efficacy. For instance, cobra venom factor (CVF) depletes the C3 component of the complement system and has been shown to facilitate OV infection116. Cyclophosphamide (CPA) has also been used to attenuate antibody- mediated activation of the complement system, serum neutralization of virus, and reduce peripheral blood mononuclear counts that are responsible for producing an antiviral cytokine response117. Due to the pleiotropic immunosuppressive nature of CPA and its ability to halt the antiviral immune response, in vivo CPA treatment significantly reduced viral clearance and increased viral propagation117,118. Additionally, CPA was able to attenuate intratumoral recruitment of macrophages, microglia, and NK cells following

OV administration while concomitantly lowering the levels of IFN-γ 90. Taken together, these findings demonstrate that targeting the immune response to OV therapy is a particularly useful modality towards achieving enhanced OV efficacy.

An additional pharmacologic approach that has been demonstrated to enhance OV therapy is the use of the histone deacetylase inhibitor valproic acid (VPA). VPA has been shown to enhance the efficacy of oncolytic HSV119 through the inhibition of IFN-I,

STAT-1, PKR, and PML signaling within infected glioblastoma cells. By targeting intracellular mediators that are responsible for creating an antiviral state120, this

12 therapeutic approach targets the antiviral host response prior to the recruitment of antiviral cellular mediators. As a result, VPA and CPA are functioning in two distinctly different ways, while creating an environment that enhances the overall potential for viral propagation.

Continued work must be undertaken to differentiate the divergent actions of CPA and

VPA on the various cellular components of the immune response (e.g. NK cells versus macrophages). Clarifying the roles for these two drugs in NK and macrophage infiltration, cytokine production, cellular activation and subsequent cellular signaling, will foster a greater understanding of the these cellular populations in the context of OV therapy. Moreover, by obtaining a greater understanding of these two agents, additional pharmacologic modalities can potentially be pursued. Lastly, while several groups are either in preparation to begin or have already started clinical trials combining OV with

CPA121, a greater understanding of the actions of VPA on OV therapy will facilitate the movement of this cotherapy into clinical trials as well.

Taken together, the overarching hypothesis of this work suggests that NK cells preferentially clear virally infected GBM. This occurs following their rapid activation and recruitment to the site of oHSV infection, thereby limiting the oHSV inoculum from undergoing multiple rounds of viral replication and disseminating through the tumor microenvironment. The corollary to this hypothesis is that pharmacologic circumvention

13 of this initial NK cell response has the potential of enhancing viral propagation within the tumor microenvironment, leading to enhanced tumor clearance.

14

Figure 1.1: Human NK cell receptors and ligands.

Human NK cell activation status and cytotoxicity is coordinated through the binding of various activating and inhibitory receptors on their cell surface to target cell ligands.

15

Table 1.1: Clinical trials using oncolytiv viruses for the treatment of malignant gliomas.

Median Genetic Route of Max OV Virus Pts. Survival Alteration Delivery Dose (mo.)

1x109 21 I.T. 6 p.f.u. Deletion of both 2 doses: HSV-1 γ34.5 copies G207 I.T. & (strain F) and disruption 1.15x109 6 I.A.B post 6.6 of ICP6/RR p.f.u. tumor

resection

1x105 9 I.T. N.I. p.f.u.

I.T. 4-9

HSV-1 days prior 1x105 Deletion of both 12 N.I. 1716 (Glasgow to p.f.u. γ34.5 copies strain 17) resection

I.A.B. 1x105 12 post tumor N.I. p.f.u. resection

continued

16

Table 1.1 continued

I.A.B. ONYX Deleted E1B 1x1010 Adenovirus 24 post tumor 6 -015 gene p.f.u. resection

Newcastle NDV- Disease None 14 I.V. 55 BIU 8 HuJ Virus

Reolys 1x109 Reovirus None 12 I.T. 5 in p.f.u.

Legend: B.I.U.: billion infection; I.A.B.: injected into adjacent brain; I.T.: intratumoral;

I.V.: intravenous; mo.: months; N.I.: not included; p.f.u.: plaque forming units; pts.: patients

17

Chapter 2: Background

Section 1—Glioblastoma overview

The World Health Organization (WHO) classifies astrocytomas on the basis of histologic features into four prognostic grades: grade I (pilocytic astrocytoma), grade II (diffuse astrocytoma), grade III (anaplastic astrocytoma), and grave IV (glioblastoma)122. The prevalence of malignant gliomas (grade III and IV tumors) is 5 out of 100,000 individuals, with glioblastoma accounting for 60-70% and anaplastic astrocytoma accounting for 10-15%8. While glioblastoma represents the most common form of primary brain tumor, it also stands out as the most devastating central nervous system

(CNS) malignancy. Histologic features of anaplastic astrocytomas include increased cellularity, nuclear atypia, and numerous mitoses. Features of glioblastoma also include areas of microvascular proliferation and pseudopalisading necrosis. Definitive diagnosis of glioblastoma requires surgical biopsy and histological confirmation of GBM’s hallmark features. The ability of glioblastoma to become highly infiltrative within the stromal tissue and its concomitant histological heterogeneity contributes both to its highly invasive nature and devastating prognosis.

Primary GBM occurs de novo in patients over 50 years old and typically include the following aberrations: EGFR amplification (approximately half of which possess the

18 constitutively autophosphorylated vIII version which lacks the extracellular ligand binding domain), loss of heterozygosity (LOH) of chromosome 10q, and deletion of both

PTEN and p16. In contrast, secondary glioblastoma manifests in younger patients as a low grade anaplastic astrocytoma that develops over time into glioblastoma. Features of this less common glioblastoma include mutations in p53, over-expression of PDGFR, abnormal p16 and retinoblastoma (Rb) pathways, and LOH of 10q. Despite these developmental and genetic differences, the morphology and outcome to standard therapy for these two forms of glioblastoma are equivalent8.

In order to achieve improved success in treating glioblastoma, it is becoming increasingly apparent that a greater understanding of the molecular mechanisms leading to this disease is needed. By deciphering the critical pathways leading to tumor formation, investigators can focus on novel targets. Until the recently completed Cancer Genome Atlas Research

Network’s (TCGA) genomic characterization of glioblastoma, our understanding of the key genetic events in glioblastoma included: 1) dysregulation of growth factor signaling through the amplification and mutational activation of receptor tyrosine kinase genes; 2) activation of the phosphatidylinositol-3-OH kinase (PI(3)K) pathway; and 3) inactivation of the p53 and Rb tumor suppressor pathways123. Additional work that has investigated genome-wide profiles has demonstrated both remarkable heterogeneity amongst glioblastoma and unique genomic profiles that allow for stratification into glioblastoma subclasses124-129.

19

The work from the TCGA and its genomic sequencing of 206 glioblastomas uncovered key genetic alterations in various signaling pathways. For instance, 88% of sequence alterations targeted the RTK/RAS/PI(3)K pathway; p53 signaling was altered in 87% of cases; and 78% of the cases exhibited Rb signaling alterations130. An additional large- scale sequencing initiative of glioblastoma has identified isocitrate dehydrogenase 1 as a mutational hotspot131. With the arrival of next-generation sequencing, uncovering the pattern of these mutations has the potential of transforming future therapies for glioblastoma. For instance, as personalized medicine with genomic sequencing becomes increasingly prevalent in the clinical setting, future trials will be able to specifically tailor therapeutic decisions based on the pattern of mutations that are associated with an individual’s glioblastoma.

Section 2—Glioblastoma and the immune environment

Aberrations in the immunogenicity of glioblastomas are an additional hallmark of this disease. Despite the presence of activating ligands for NK cells and co-stimulatory T-cell ligands132, patients exhibit deficiencies in cellular immunoreactivity133 and a shift towards a Th2 cytokine profile134-137. This results from the production of a variety of released inhibitory signals including TGF-β138,139, IL-10140, HLA-G141, and B7-H1142.

There is also a concomitant reduction in the production of TNF-α and IL-12108. Recent work by Wu et al.143 suggests that glioma cancer stem cells may be mediating this immunosuppressive profile by secreting macrophage inhibitory cytokine-1 (MIC-1),

TGF-β, and soluble colony-stimulating factor (sCSF). This results in macrophage

20 recruitment and the adoption of a tumor associated/immunosuppressive M2 macrophage phenotype that is STAT3 dependent143. Interestingly, STAT3 inhibition was able to reverse this M2 phenotype in vitro, suggesting that it is a potential therapeutic target that deserves further investigation143.

In addition to alterations in the production of inflammatory mediators, glioblastoma has been associated with decreases in overall numbers of T-lymphocytes, helper T cells, cytotoxic T cells, B-lymphocytes, and reduced fractions of CD4+/CD8+ cells108.

Interestingly, in this later study there was no difference in the number of circulating NK cells108. Lastly, patients with more aggressive tumors based on WHO classification exhibited a greater impairment in immunity108. Taken collectively, achieving successful antitumor immunity will require a strategy consisting of either antagonizing tumor mediated immune suppression or the administration of robust pro-inflammatory mediators that can circumvent the pervasive immunosuppressed tumor microenvironment.

Section 3—Oncolytic viruses as a therapeutic modality

For almost two decades, there has been interest in using viruses to deliver genes into cells144. One particular approach consists of oncolytic viruses (OV), which can selectively enter and replicate in neoplastic cells leading to their lytic destruction with minimal damage to surrounding normal tissue. OV include a wide range of viruses that have been selected or genetically engineered such that viral replication is limited to

21 permissive cancer cells with specific mutated cellular pathways. OVs have been designed to replicate only in tumors that either have activation of specific oncogenes or inactivation of specific tumor suppressor pathways145-151. Some OVs demonstrate selective tropism for entry into tumor cells117,152-155. Second generation viruses that are

“armed” by incorporation of pro-drug activating genes42,109,156-161, imaging genes162,163, immunostimulatory genes87,164-167, and anti-angiogenesis genes168,169 are currently being investigated for safety and efficacy. Table 2.1 presents an overview of specific OVs along with their salient properties that are being studied for the treatment of malignant gliomas.

The appropriate route of delivery of OV remains to be defined in terms of advantages and disadvantages. For example, intratumoral viral delivery has the advantage of circumventing rapid viral clearance within the bloodstream due to antibody and complement neutralization of the virus, clearance by the liver, viral binding to non-tumor cells that contain receptors for the virus, and barriers to migration across the vascular endothelium. However, intravenous administration is the route of choice for the treatment of both primary invasive tumors and metastatic disease. Methods of avoiding these limitations to viral administration include the development of various stealth agents and carrier cells to achieve non-immunogenic viral delivery.

With China’s recent approval of the first oncolytic virus, adenovirus H101170, a number of clinical trials are underway in the United States and Europe. Table 1.1 presents a

22 summary of the glioma clinical trials that have been performed to date69,79-85. Through the process of testing OVs in the clinic, however, a number of questions must be addressed. For instance, the pharmacokinetics of viral infection, replication, and spread should be ascertained non-invasively. Two novel oncolytic measles viruses are attempting to answer these questions. First, a measles virus encoding the soluble extracellular human carcinoembryonic antigen (CEA) allows for noninvasive analysis of viral propagation by measuring CEA levels171,172. Second, by incorporating the thyroidal sodium iodide symporter in the measles vector, clinicians are able to use radioactive iodine tracers in order to monitor the status of viral infection using single-photon- emission computed tomography or positron-emission tomography162,173-175. Beyond questions related to pharmacokinetics, clinical implementation of OV is hampered by technical challenges in producing large amounts of high titer virus. Lastly, performance of phase III clinical trials to assess clinical utility and guide the future directions of basic research in the field of OV therapy is needed.

For instance, preclinical data suggests that the ability of OVs to amplify within cancer cells should lead to increased intratumoral titers independent of the initial inoculum87-

89,145,176-178. While these findings have been corroborated by numerous in vitro findings, clinical efficacy has been limited due to significantly attenuated in vivo viral replication102,179-185. In fact, a recent clinical trial shows replication of inoculated virus in tumor, albeit at levels that appear to be fairly reduced80,186. Attenuated in vivo viral replication may be due to inefficient intratumoral viral dispersal, to barriers imposed by

23 the tumor microenvironment, and/or by rapid viral clearance by host immune responses.

Future clinical trials will need to take these host factors into account in order to achieve maximal OV-mediated tumoricidal activity while simultaneously avoiding systemic toxicity to the host. Elucidation of a variety of tumor- and host-based factors that limit viral infection, replication, and propagation, could lead to the design of combinatorial molecular approaches aimed at pairing oncolysis with pharmacologic agents designed to circumvent such host barriers to OV lysis of tumors. Additionally, certain classes of pharmacological agents can alter cellular homeostasis and activate cellular cascades that provide an environment conducive for viral replication.

Section 4—The immediate host response following OV infection

A major assumption in the area of OV therapy of tumors has been that even a small initial dose of a replication competent OV will amplify through successive rounds of viral replication, resulting in eventual infection and eradication of the entire tumor144. In reality, however, viral distribution appears stunted and viral yields within tumors actually decrease as a function of time145,186. In order to fully explain this, one needs to evaluate how oncolytic viruses function within the context of the tumor microenvironment; in fact, it is critical to understand what is required for effective viral-mediated tumor killing and what could limit viral replication.

Tumor clearance by OV is an unlikely single modality due to limitations imposed within the tumor microenvironment187. Sequential steps predicted to occur during OV killing

24 include: a) infection of individual cells, replication within them, and subsequent cell death; b) induction of an adaptive anti-tumor immune response triggered by viral infection; and c) stimulation of localized inflammation. A consequence of the immunosuppressive nature of certain tumor types, particularly glioblastomas, has required investigators to create novel methods for modulating host immunity in order to achieve potent anti-tumor immune responses. For instance, the lack of immunostimulatory signaling present on intracranial glioblastoma cells was circumvented using an oncolytic herpes simplex virus (HSV) vector expressing IL-4. This virus was able to mediate antitumor efficacy not only through viral oncolysis but also by induction of a CD4 and CD8 antitumor response187. An additional barrier towards achieving significant antitumor immunity in the context of OV is the ability of certain viruses, such as HSV-1 through its ICP47 protein, to circumvent the host response by binding to transporter associated with antigen presentation (TAP) and blocking MHC-I mediated antigen presentation following viral infection188-190. In order to address this limitation, a novel HSV was produced that lacked the gene encoding ICP47 and it was found to significantly enhance the stimulation of tumor-infiltrating lymphocytes in vitro191. Since

ICP47 only recognizes human TAP, the authors conclude that evaluation in patients will be needed to fully assess the safety and efficacy of this novel oncolytic virus. Taken collectively, through a combination of viral tumor clearance and immune modulation, a

“perfect storm” of tumor killing could potentially occur: replication of the virus in a dying tumor; secretion of pro-inflammatory cytokines; and recruitment of inflammatory cells into the tumor microenvironment leading to tumor cell destruction103,187,191,192.

25

However, while viral replication can work synergistically with the immune system to elicit tumor killing, the immune system also functions as a “double-edged sword”. Upon viral infection, a series of antiviral mechanisms are activated via the initial innate immune response to the virus. This response has been demonstrated to limit successful viral propagation and tumor clearance15,193-196. Consequently, these barriers must be circumvented to achieve viral replication and the subsequent activation of an adaptive antitumor immune response146. These barriers consist of: intracellular signaling and antiviral defenses15,20,150,197, extracellular tumor environmental barriers20,103,198, and the active host response to ongoing oncolytic virotherapy102,103,116,117,145,181,195,199-204 (Figure

2.1). In this section we will overview the effect of each of these barriers on oncolysis.

Section 4.1—Barriers to viral entry

Viral entry has represented an obstacle within the field of oncolytic viral therapy. For instance, low levels of adenovirus receptors on tumor cells have posed a hindrance for initial oncolytic adenoviral binding and entry into tumor cells. As a result, extensive manipulations of receptor binding proteins and the use of chimeric serotypes have been required in order to achieve enhanced anti-tumor efficacy205. Measles virus enters human cells through two known receptors: CD46, a regulator of complement activation, and the signaling lymphocyte activation molecule (SLAM), which is expressed on activated B and T cells and macrophages206,207. The use of oncolytic measles virus in the treatment of glioblastoma has been hindered by the ubiquitous expression of CD46 in normal brain

26 and blood-brain barrier208,209 and the interaction with SLAM, which has been associated with immunosuppression after infection with wild-type virus206. In order to circumvent this receptor barrier, Allen et al. constructed a human IL-13 displaying retargeted measles virus with ablated natural viral entry through CD46 and SLAM210. This virus displayed enhanced specificity for glioblastoma since nearly 80% of these tumors display the IL-13 receptor211-213. By retargeting measles virus against tumor specific receptors, novel viruses will have improved viral specificity and circumvent the clinical challenge of variable CD46 expression. In addition to measles virus, this approach of receptor retargeting has also been tested with oncolytic HSV (oHSV) targeted to either HER-2214 or EGFRvIII215 expressing tumors.

For oHSV infection, receptor expression has generally not been considered a significant challenge for the development of oHSV-based therapy. However, studies investigating the abundance and requirements of HSV receptors across multiple tumors are lacking.

Initial HSV binding occurs through the interaction between the viral surface glycoproteins, gB and gC, and heparan sulfate which is relatively ubiquitous on most cells. After this initial attachment, viral entry occurs through the binding of gD to one of multiple Herpes virus entry (Hve) mediators. HveA (HveM) and HveC (nectin-1 or

CD111) are the two main receptors involved in this process; however, HveB (nectin-2 or

CD112) may also be used in some cases, such as with laboratory strains that have specific gD mutations216. Lastly, 3-O sulfated heparan sulfate is also a receptor for gD217. While nectin-1 appears to be the most efficient receptor, total receptor density directly correlates

27 with virus entry efficiency218. Future studies are needed to study whether HSV receptor expression profile of various tumors correlates with HSV infectability. This would have the potential of linking a patient’s receptor biomarker with predictive success of oHSV virotherapy.

Section 4.2—The intracellular antiviral response

Pattern recognition receptors have evolved to detect invading pathogens and they fall into two broad categories: toll-like receptors (TLR) and RIG-I-like helicases (Figure 2.2).

TLR are abundantly expressed on plasmacytoid dendritic cells (pDC), are found on either cell surfaces or endosomes where they detect a variety of pathogen associated molecular patterns (PAMP) 219, and transmit their downstream signals through their cytoplasmic

Toll/IL-1 (TIR) domain. The response of ligand binding to a TLR depends on the TIR adapter protein that is associated with each TLR. With the exception of TLR-3, myeloid differentiation primary response protein 88 (MyD88) associates with the TIR domain of each TLR and ultimately leads to the downstream activation of nuclear factor-kappa B

(NF-κB) and the production of various inflammatory cytokines, including TNF-α, IL-6, and IL-1β. TLR3, however, recognizes double stranded RNA (dsRNA) and rather than signaling through MyD88, it associates with Toll/interleukin 1 receptor domain- containing adaptor protein inducing interferon β (TRIF) to activate both NF-κB and interferon regulatory factor 3 (IRF-3). IRF-3 activation leads to its translocation into the nucleus and the induction of IFN-β expression220,221.

28 pDC function as the host’s professional interferon (IFN) producing cells due to their TLR expression, their ability to actively produce IFN-I, and their critical role in limiting viral infection222. These cells are found in the blood, peripheral lymphoid organs, cerebrospinal fluid, choroid plexi, and brain meninges; however, they are not found in non-inflammed brain parenchyma. Their presence in the parenchyma is only achieved either experimentally, following administration of Fms-like tyrosine kinase 3 ligand in the brain223, or in states of chronic neuroinflammation224-226.

RIG-I is a critical mediator of IFN production and viral clearance in the majority of cell types, including fibroblasts, epithelial cells, and conventional dendritic cells222. This pattern recognition receptor is ubiquitously expressed, located in the cytosol where it detects dsRNA that is unique to virally infected cells, and signals through the mitochondrial membrane associated interferon promoter stimulator 1 (IPS-1) adaptor protein. Once dsRNA binds to RIG-I, downstream signaling events reach IPS-1 and branch out into either NK-κB, IRF-3, or IRF-7 activation. IRF-3 is a constitutively expressed protein that shuttles between the cytoplasm and nucleus. Once IRF-3 is phosphorylated at its C-terminus, it remains localized in the nucleus where it serves as a transcription factor for IFN-β, IFN-α1, and RANTES227,228. While IRF-7 also requires a

C-terminal phoshorylation event to become activated229,230, it differs from IRF-3 in several ways: it is only constitutively expressed in B cells and DC while its expression elsewhere only occurs following viral infection or IFN induction; it has a half life of 30 minutes229,231,232; and it actively transcribes IFN-α4, 7, and 14 228.

29

As the host has a variety of methods for detecting invading pathogens and ultimately producing IFN-I, this underscores the importance of IFN-I production as an antiviral mediator. For instance, experiments that have depleted the IFN ligand demonstrated the necessity of IFN-I in abrogating initial viral replication99. While IFN-I is not intrinsically antiviral, its production is able to induce a variety of changes through binding to nearby

IFN receptors leading to the downstream activation of IRF93-95 and interferon-stimulated genes (ISG) that are responsible for creating an antiviral state94,96-98.

In addition to the PAMPs previously described, the double-stranded RNA-activated protein kinase (PKR) is activated following IFN-I production 149. PKR recognizes foreign and abnormal nucleic acid structures that accompany viral infection233. Binding of PKR to a dsRNA leads to an activated form of PKR that has the ability to phosphorylate eIF2a and halt cellular protein synthesis. Through PKR induced abrogation of protein translation, the host cell is no longer able carry out viral protein production.

Most viruses have evolved mechanisms to counter intracellular defense responses.

Interestingly, antiviral defenses are often disrupted in tumor cells. This can provide researchers with stratagems for engineering attenuated OVs that can selectively replicate only in cells that lack antiviral defense response. For example, vesicular stomatitis virus

(VSV)150, reovirus152, and myxoma virus153 are naturally sensitive to IFN so their

30 replication is selective for tumor cells where this pathway has been reported to be defective. Similarly, reovirus, VSV, and oncolytic HSV-1 have been reported to selectively replicate in tumor cells with an activated Ras/MEK pathway, which can counter the activation of antiviral PKR in cells24,234-236. However, evidence exists that despite altered IFN signaling pathways in certain tumors, oncolytic viruses are still subject to control by the innate immune defenses of human tumor cells 237. As a result, specific approaches aimed at circumventing these antiviral defenses may lead to enhanced viral replication and spread.

Section 4.3—Extracellular tumor microenvironment barriers

The extracellular tumor microenvironment (ECM) or “the cancer field” consists of secreted proteins, proteases, growth factors, stromal and immune cells and tumor vasculature. The significance of the ECM in governing tumor growth and also its response to therapy is increasingly being appreciated. Recent studies investigating the complex interactions between OV therapy and tumor ECM have uncovered the highly significant impact of the tumor microenvironment on oncolysis. After viral replication and lysis of the infected cell, the progeny OV resulting from the “virus burst” have to spread from one infected cell to the next. The extracellular tumor microenvironment consists of secreted proteins and proteoglycans which form an inhibitory scaffold limiting the spread of OV particles within the solid tumor238. Apart from the physical inhibition, the acidic tumor microenvironment and high interstitial tumor pressure present additional obstacles for viral propagation and spread in the tissue 53,239-241 (Figure 2.3).

31

Following wild-type viral infection, a classical physiologic response is vasodilation and hyperpermeability100,101. This places blood vessels as an integral intermediary in the inflammatory host response to infection242. During inflammation, peripheral leukocytes and monocytes extravasate into tissue by initially adhering to endothelial cells that line the vascular walls, ultimately leading to endothelial activation. This activation leads to subsequent hyperpermeability in the vascular walls and increased tissue edema, ultimately enhancing perivascular inflammatory cell infiltration242,243. This vascular leakage can become detrimental to OV replication and spread within the tumor204, thereby accentuating an additional factor that must be addressed in order to achieve successful OV therapy. Various approaches have been examined to enhance virotherapy by stabilizing tumor vasculature, reducing the neovascular response, and reducing the inflammatory cellular infiltrate199,204.

Section 4.4—Immune responses to oncolytic viruses and its cellular mediators

Likely the most significant limitation to virotherapy is the active innate immune response to the virus that can occur fairly rapidly after OV infection. The innate immune system provides an initial potent line of defense that limits initial viral infection, replication, and spread; signals for the maturation of antigen presenting cells; and activates the cellular components of the adaptive immune system. The importance of this concept has been elucidated in several models, including VSV, wherein initial intratumoral viral replication is followed by a dramatic decline in viral titers over the following days17. Since anti-viral

32 antibodies were not produced until 5 days post-OV, the innate immune system (including granulocytes, natural killer (NK) cells, NKT cells, and macrophages) that is recruited to the site of infection is considered a major player in limiting viral propagation244.

Depletion of mononuclear cells145 or antiviral cytokine mediators such as IFN-γ202 has been shown to cause a significant increase in intratumoral viral titers and anticancer effects.

While neutrophils are the first antiviral responders that are recruited to a site of infection, efficient viral clearance at the cellular level requires both NK cells and monocyte-derived cells. Activated NK cells109 mediate direct lysis of infected target cells by releasing cytotoxic granules containing lytic enzymes or by binding to apoptosis-inducing receptors on target cells21. NK cell-mediated preferential lysis of HSV- or vaccinia-virus infected cells has been shown to prevent viral dissemination to neighboring cells245.

While recruitment of NK cells to infected tumor tissue has correlated with reduced viral spread and OV efficacy, IFN-γ production by NK cells has also been shown to set the stage for the subsequent adaptive immune response246,247.

Apart from NK cells, macrophages also play a critical role in OV clearance. Upon viral infection, resident or recruited macrophages initially secrete IL-12 to activate NK cells while NK cells complete the feedback loop by secreting IFN-γ—the prototypic macrophage activator, without which macrophages cannot clear microbes246 (Figure 2.4).

In fact, recruitment of infiltrating monocytic cells has been shown to coincide with

33 clearance of over 80% of HSV derived oncolytic viral particles 102,202,45. Increased intra- tumoral presence of macrophage/microglia cells has also been reported in human patients treated with adenovirus38,248 or HSV1-derived OV249 indicating the global significance of macrophages in OV therapy.

It is important to note that while the OV-mediated induction of an antiviral inflammatory state is thought to be detrimental towards oncolysis, a recent study indicated that it could also contribute to tumor killing. During inflammatory reactions, activated neutrophils adopt a “rigid” phenotype which can result in clogging of small capillaries156. Systemic delivery of VSV and vaccinia virus has been shown to initiate a robust recruitment of neutrophils from the vascular system into the tumor. This robust immune cell infiltration resulted in neutrophil-dependent initiation of microclots within tumor blood vessels leading to a “choking of the blood vessels”. The ensuing disruption of perfusion and increase in tumor hypoxia induced tumor cell apoptosis and contributed to tumor cell killing103.

While neutrophil-mediated choking of tumors may be beneficial, antibody-mediated neutrophil depletion facilitated extensive viral replication and spreading throughout the tumor103. Additional studies with a variety of tumor models and oncolytic viruses will need to be performed in order to determine if this mechanism of tumor cell death and inhibition of viral spread is dependant on tumor type, virus type, or route of OV administration.

34

In addition to the cellular components of innate immunity, the role of pre-existing immunity must be considered in the context of OV therapy. With the prevalence of preexisting immunity to many oncolytic viral vectors, particularly HSV and measles, its role as a potential barrier must be considered. Multiple studies have examined the role of preexisting immunity to HSV in the oHSV treatment of colon cancer (intratumoral injection)250 and liver metastases (intravascular injection)251. Interestingly, in both studies previous exposure to HSV did not limit the efficacy of oHSV therapy. These findings provide preliminary evidence that previous immunity to the oncolytic viral vector has limited impact on the success of virotherapy; however, future studies are needed to extend these findings across multiple viruses, tumors, and routes of OV administration.

Despite their antiviral properties, neutrophils and NK cells have pleiotropic effects that may also be critical in tumor killing. For instance, neutrophils, in addition to CD8 T cells, have been shown to contribute to HSV179, VSV103,180, and measles virus 177,252- related virotherapy efficiency. Similarly, NK cells have been shown to augment the tumoricidal effects of oncolytic HSV. In a melanoma model, NK cells have been defined as an essential cellular component for VSV efficacy180. In this model, NK cells functioned synergistically with the adaptive antitumor immune response, launched in response to viral antigens expressed by tumor cells. Therefore, it appears that NK cells can serve a dual function—both as potential inhibitors of viral replication and as critical

35 mediators to establish an effective antitumor immunity following viral antigen presentation within the tumor cells. These findings emphasize the impact of variations in tumor models, anatomical location of the tumors, and properties of the viruses that are being tested253.

In future studies, a refined approach will be needed to manipulate individual cell populations while considering both the timing and nature of the intervention in order to maximize therapeutic regimens. One promising, albeit simplistic, approach will be to combine OV inoculation with initial transient immunosuppression in order to achieve initial viral replication, followed by restoration of immune activity to harness the immunotherapeutic potential of virotherapy.

Section 5—Immune response to HSV infection

Following viral infection, HSV-1 and the host each posses various interacting factors that attempt to achieve viral replication and block foreign pathogens, respectively254. HSV-1 achieves cellular entry through initial binding of gB and gC to heparan sulfate moieties of cell surface proteoglycans255 with subsequent gD mediated viral fusion to the target cell256. After attachment and fusion, the capsid and its associated tegument are transported to the nuclear pore through the microtubular network. In order to circumvent the initial antiviral IFN-I response, HSV-1 encodes various infected cell proteins (ICP).

These include ICP34.5, ICP0, and ICP27. ICP34.5 inhibits activated protein kinase R signaling while ICP27 attenuates Jak-STAT signaling. By targeting these signaling

36 pathways, HSV is able to avoid protein translational arrest. ICP0 inhibits the nuclear accumulation of IRF3 and IRF7, thereby circumventing IFN production and subsequent cellular antiviral defenses257-262.

Following viral infection, a robust immune response in the host is elicited through the recognition of invariant viral pathogen associated molecular patters (PAMPs) by toll-like receptors. TLR-2, 3, and 9 have all been implicated in this host response to HSV-1 infection. TLR-2 is found on the cell surface of microglia and astorcytes263. Through either MyD88 or TIRAP signaling cascades, NF-κB is activated and a variety of pro- inflammatory cytokines are elaborated including TNF-α and IL-15264-268. TLR-2 has also been shown to work in concert with TLR-9 to control HSV-1 infection through NK cell recruitment in the brain to decrease viral loads269,270. TLR-9 recognizes unmethylated

CpG DNA motifs and is found within the endosomes of microglia and astrocytes.

Stimulation of this signaling cascades results in the rapid production of IFN-I in an

IRAK-4 and MyD88 dependent manner271-274. TLR-3 is found in cell compartments of microglia, astrocytes, oligodendrocytes, and neurons. Double stranded HSV-1 and non- transcribed RNA intermediates binds to TLR-3 leading to downstream activation of IRF-

3 and NF-κB in a MyD88 independent fashion266,267,275-278. While NF-κB activation is the final product of multiple TLRs, IRF-3 is mobilized by only TLR-3 and TLR-4, leading to the induction of a unique set of genes critical for antiviral defense276.

37

Following viral induced IFN-I production, Jak-Stat signaling is heightened in addition to induction of RNAse L, PKR, and Mx protein GTPases279-281. Between the ability of

RNAse L to degrade intracellular viral mRNA and PKR mediated protein cessation, viral replication is significantly attenuated. IFNs also facilitate the maturation of DCs and activation of NK cells, DCs, B and T lymphocytes282.

Section 6—Intracellular pathways affecting virotherapy

The metabolic/replication potential of a cell has a tremendous impact on OV replication.

Intracellular changes in cell signaling cascades can transform a cell into a host that encourages or discourages OV propagation. For instance, the cellular stress response, induced by pharmacological treatment, results in a variety of changes that have significant impact on viral infection and dissemination. These can range from alterations in protein expression to changes in cell cycle status.

The acquisition and conservation of cellular genes by wild-type viruses for specific tasks has been historically documented283,284. A defining feature of OV therapy adopts a similar approach whereby genetically engineered viruses frequently lack a specific gene whose function must be provided by the host cell in order to achieve successful viral propagation. An example of such a gene encoded by HSV-1 is γ34.5. The protein product of this gene, ICP34.5, precludes the shutoff of host protein synthesis and premature cell death285. Notably, however, the carboxyl terminus of ICP34.5 has significant homology to the carboxyl terminus of mammalian growth arrest and DNA

38 damaging inducible protein (GADD34)285, a cellular stress protein that circumvents apoptosis by suppressing cell division during DNA repair286-288. GADD34 recruits protein phosphatase-1 and dephosphorylates the inactivated mRNA translation initiation factor eIF2a allowing for viral protein synthesis to occur. In the context of oHSV therapy where the viral γ34.5 gene is frequently deleted to limit unintended virulence, the function of ICP34.5 appears to be provided in trans by GADD34288. Taken together, identifying ways of enhancing GADD34 induction in the presence of HSV lacking γ34.5 may provide a useful strategy for enhancing virulence within the tumor targets. A similar type of engineering is seen by the finding that an activated MEK pathway in cells can substitute for the lack of γ34.5 function and allow robust replication of the γ34.5 mutant HSV-1 in vitro and in vivo48,289.

Section 7—Pharmacological modulation of host factors to enhance OV therapy

The growing body of literature on the limitations induced by the various intra- and extracellular host defense responses to OV therapy has led to the development of several strategies to combat these undesirable changes to enhance tumor oncolysis. Exploiting pharmacological agents to manipulate cancer cells and their microenvironment to enhance OV therapy is a promising approach to achieving improved OV efficacy. Results from the preclinical testing of several pharmacological drugs that target various aspects of the immune response used in combination with OV therapy have revealed the potential of this strategy to synergize with OV therapy. The following sections will include various pharmacologic approaches that have been shown to augment virotherapy

39

Section 7.1—Immune modulators

Recent studies investigating the impact of combating the antiviral host immune responses with pharmacologic agents has led to the identification of several drugs that synergize with OV therapy. The effect of CVF mediated depletion of serum complement proteins,

CPA mediated depletion of peripheral blood mononuclear cells (PBMC), and clodronate liposome (CL) mediated exhaustion of phagocytic cells have been shown to increase OV persistence.

The complement system consists of a series of serum proteases that result in the destruction of virions/infected cells through a number of routes, including the formation of membrane attack complexes on the surface of infected cells and enveloped viruses; production of anaphylatoxins that recruit additional immune mediators to the site of infection; phagocytosis of opsonized virions and infected cells; and the direct neutralization of virus following complement binding to the virion surface290. Therefore, drugs that can temporarily inhibit complement could provide a therapeutic advantage to

OV therapy. CVF is the prototypical complement inhibitor that depletes the C3 component of the complement system. In fact, in vivo depletion of complement by systemic administration of CVF has been shown to facilitate OV infection116. However the benefits of CVF, are short-lived, as there was no evidence of increased OV persistence in tumors after infection 117.

40

Upon antigen recognition, the Fc region of antibody binds complement C1 and activates the complement cascade. Treatment of animals with CPA has been shown to reduce the serum neutralization of virus, partly due to reduction in IgM and anti-HSV antibody levels in treated animals118. In contrast to CVF, treatment of animals with CPA prior to

OV therapy reduced viral clearance and increase viral propagation in vivo117,200. This translated into increased cancer cell killing in vivo even at very low doses of OV 182. CPA is also a DNA alkylating agent leading to DNA damage and tumor cell apoptosis117.

However in vitro treatment of glioblastoma cells with 4-hydroperoxyCPA (the activated form of CPA) did not increase OV replication, indicating that the observed augmentation in OV efficacy was not a direct effect of CPA on viral replication. The increase in therapeutic efficacy has been attributed to CPA mediated reduction in PBMC counts that can limit the antiviral cytokine response, ultimately contributing to the enhanced anticancer efficacy117. This is corroborated by findings showing diminished intratumoral infiltration of macrophages/microglia and NK cells and lower levels of IFN-γ in glioblastomas treated with OV and CPA202. Collectively, these findings indicate that

CPA mediated improvement in OV efficacy is a product of the immunosuppressive action of CPA rather than synergistic cell killing between CPA-mediated cellular apoptosis and OV-mediated lytic destruction of cancer cells.

Among the pleiotropic immunomodulatory effects associated with CPA, lower doses of

CPA have also been shown to enhance the immune response against tumors291-293 by transiently depleting regulatory T cells (Tregs) that suppress NK cells and antitumor CD8

41

T cells291-298. In order to achieve this biphasic response, it will be critical to establish a dosing schedule for modulating the different phases of the immune response. This has the potential of allowing for an initial enhancement of viral oncolysis followed by the production of a delayed immuno-enhancing suppression of Tregs, a step that is critical for the later adaptive immune response and vaccine-like effect against the tumor146. As a result, CPA is impacting the tumor microenvironment by limiting both the influx of antiviral cellular mediators and the antiviral cytokine milieu, hence setting up the stage for reduced viral clearance and maximizing oncolysis.

Apart from HSV-1 derived OV, CPA has also been shown to increase the oncolytic capacity of other OVs derived from HSV-2183, adenovirus102, and reovirus299,300. Based on the very promising preclinical results seen with CPA and OV, the combination of

CPA with measles virus is currently being evaluated for safety and efficacy in human patients121.

More recently CLs have been used to investigate the importance of macrophages in OV clearance in vivo. Clodronate encapsulated in liposomes is engulfed by phagocytic cells resulting in intracellular accumulation of apoptosis inducing clodronate184. CL-mediated depletion of peripheral phagocytic cells resulted in a 5-fold increase in OV titers in intracranial glioblastoma. While these findings partly recapitulated the effect of CPA on

OV replication, they were unable to achieve the enhanced survival demonstrated with

CPA145. A potential reason for these findings may relate to the inability of clodronate to

42 cross the blood brain barrier, thereby limiting its ability to deplete phagocytic microglial cells in addition to peripheral macrophages.

Section 7.2—Histone deacetylase inhibitors

Histone acetylation/deacetylation is a major factor in regulating chromatin structural dynamics during transcription. Histone deacetylase inhibitors (HDACi) have been shown to induce cellular apoptosis, exert antiangiogenic activities, and also interfere with transcriptional activation of antiviral genes after IFN stimulation or virus infection185,301-

303. They are currently being pursued as potential anticancer agents304-307 308,309 alone and in conjunction with chemotherapy310-312. HDAC activity is critical for IRF-3 gene expression in virus-infected cells301 and its inhibition can prevent the transcriptional activation of ISG in response to viral infections301,303,313-319. Given the strong antiviral and antitumorigenic effects of HDACi, they are currently being investigated as potential agents to modulate OV efficacy. We will discuss the use of valproic acid (VPA) and trichostatin A (TSA) in conjunction with OV therapy.

VPA is an inhibitor of HDAC, and is clinically used as an anticonvulsant and mood- stabilizing drug. VPA has also been shown to have anticancer effects in animal models and is currently being evaluated as an antineoplastic agent for several human malignancies. Apart from its direct anticancer effects, treatment of glioblastoma cells with VPA has been shown to enhance the oncolytic efficacy of oncolytic HSV-1119. This has been attributed to VPA mediated inhibition of IFN-β and IFN-mediated proteins

43

STAT1, PKR, and PML in infected cells119. The significance of this finding is heightened since STAT1 is a key transcription factor that mediates IFN signaling and its activation is responsible for establishing an intracellular antiviral state120.

TSA is a promising HDACi that functions as a potent inhibitor of cyclin D1 and arrests cell cycle progression320-322. Similar to VPA, treatment of cancer cells with TSA, in combination with OV therapy has also been shown to increase oncolysis. Combination of

HDACi with VSV in a variety of cancer cells enhanced antitumor efficacy primarily by

TSA mediated increase in mitochondrial depolymerization and cleavage of caspases 3 and 9185. Enhanced antitumoral and antiangiogenic effects of TSA in conjunction with oncolytic HSV have also been reported323. However unlike VPA, TSA treatment did not affect the IFN response and the observed synergistic killing has been attributed to enhanced degradation of cyclin D1 and VEGF inhibition323. Reduction in VEGF expression by TSA may also contribute to enhanced OV efficacy. TSA treatment of cancer cells has been shown to up regulate expression of cell surface receptors that are critical mediators of adenoviral cell entry: coxsackie-adenovirus receptor (CAR) and alphavintegrins. Consistent with this, TSA has been shown to enhance anti-tumor efficacy of conditionally replication competent adenovirus in glioblastoma cells 324-326.

Considering the diversity of cellular pathways that are targeted by HDACi, it is not surprising that studies evaluating the effect of these drugs in conjunction with OV have uncovered a variety of cellular effects contributing towards oncolysis with different OV.

44

Future studies will elucidate critical cellular pathways that should be targeted in order to enhance OV therapy.

Section 7.3—Antiangiogenic agents

Increased angiogenesis is one of the hallmarks of solid tumor growth and has been shown to be an essential prerequisite for cancer growth. Changes in the tumor “secretome”

(secreted proteins) after OV therapy have been shown to disrupt the homeostasis maintained between angiogenic and angiostatic factors resulting in increased growth of blood vessels after OV therapy199,204,327. Antiangiogenic agents are therapeutic drugs that destroy tumor vasculature resulting in increased hypoxia. While hypoxia mediated

“choking” of cancer cells has antitumor efficacy, hypoxia has also been shown to induce intracellular changes which support viral replication328. Apart from direct effects of hypoxia, increased vascularity is associated with an enhanced inflammatory response suggesting that antiangiogenic agents can be used to reduce antiviral inflammation in tumors. Thus, antiangiogenic agents have been investigated as a potential avenue for reducing the antiviral state in the tumor microenvironment and improving both OV infection and replication.

cRGD is a cyclic RGD peptide that was originally identified as antagonists for the integrins αvβ3 and αvβ5 329. These integrins are overexpressed in proliferating cancer cells and tumor endothelium330, and their interaction with the extracellular matrix mediates various intracellular signals involved in adhesion, migration, and proliferation.

45 cRGD has been shown to function as antiangiogenic factors that induce endothelial cell death and disrupt the enzymatic activity of matrix metalloproteases (MMPs)331. In preclinical studies, cRGD has been found to have significant antitumor efficacy in the treatment of glioblastoma in animal models332, and is currently being evaluated in clinical trials for efficacy in human patients. cRGD has also been shown to limit leukocyte recruitment to synovial sites of chronic inflammation333, reduce myeloid cell adhesion, and transendothelial cell migration334,335.

Its promising activity as an anti-angiogenic and anti-neoplastic agent combined with its role as an anti-inflammatory agent, suggested that it would enhance OV efficacy.

Kurozumi et al. have tested this hypothesis in a syngeneic rat glioma model91. Consistent with its known function, treatment of animals with cRGD led to a significant reduction in the number of blood vessels and reduced OV induced vascular permeability in vivo204.

Notably, cRGD pretreatment also participated in limiting OV induced pro-inflammatory cytokine profile, including IFN-γ and INF-γ induced proteins, such as CXCL9 and

CXCL11, in vivo. This reduction in OV induced inflammatory cytokine expression was accompanied by a decrease in infiltrating CD45 leukocytes204, and increased OV propagation in vivo. More significantly, cRGD, administered to animals prior to OV therapy was able to significantly enhance therapeutic efficacy of OV in animals with intracranial tumors204. Future studies will be needed to elucidate the impact cRGD on the interplay between its anti-angiogenic and anti-inflammatory responses, and whether other anti-angiogenic drugs can recapitulate the findings of cRGD when administered with OV.

46

Section 8—Placing NK cell biology in the OV context

Although a corpus of evidence has delineated both the role of NK cells in tumor clearance for various tumor models and their role in herpes virus eradication, there have been no studies to date that have accurately tracked the NK response to OV therapy, examined the NK developmental or activation status following OV infection, elucidated the signaling components that are uniquely implicated in OV clearance, or placed the NK response into the greater context of subsequent macrophage activation. As a result, it is important to understand the basic properties of NK cell biology in order to ultimately decipher their role in the context of OV therapy.

Human NK cells are divided into separate CD56bright and CD56dim populations that differ in their functional capacity and localization336. Approximately 90% of circulating and splenic NK cells are CD56dimCD16+, express perforin, and possess cytotoxic capacity when interacting with target cells337. In contrast, CD56brightCD16- NK cells are detected in lymph nodes and tonsils, lack perforin, and readily produce cytokines such as IFN-γ in response to stimulation with IL-12, -15, and -18338,339.

In mice, NK cells have been differentiated into three subsets according to CD11b and

CD27 expression340. NK cell differentiation in mice occurs from a relatively immature

C11bdullCD27+ state, to the double positive CD11b+CD27+, and ultimately to the senescent CD11b+CD27dull. Notably, both the double positive and senescent NK cells

47 have been demonstrated to secrete IFN-γ and carry out cell mediated cytotoxicity. While investigators have identified differential anatomical distribution for each type of NK cell in wild type mice, there have been no examinations into the maturation state of NK cells within the tumor microenvironment either in the absence or presence of OV.

Unfortunately, the lack of CD56 expression on mouse NK cells, combined with the difference in CD27 and CD11b staining on human and mouse NK cells limits our ability to extrapolate findings from our murine models to humans341. However, a basic understanding of the role of NK cells in response to OV therapy in mice establishes a framework for future studies in clinical trials.

While certain NK cells are sufficiently mature to produce both cytotoxic and cytokine responses, these two functions are products of the cytokine microenvironment. For instance, IFN-I, IL-12, and IL-18 are critical for the induction of NK activation342.

Moreover, much like T cells require “priming” for full activation, IL-15 has been elucidated as a cytokine that is critical for the priming function of murine NK cells341,343.

While initial efforts have examined the role of certain activating cytokines within the tumor microenvironment following OV therapy91,117, additional efforts must be undertaken to elucidate their differential role in mediating OV and tumor clearance through IFN-γ production and cellular cytotoxicity.

NK cells are able to carry out their diverse repertoire of activities through a detection system that relies on the engagement of a variety of cell surface activating and inhibitory

48 receptors on NK cells. Through the binding of these receptors, a dynamic equilibrium is achieved that differentiates the recognition of “self” cells from transformed target cells.

The activating NK cell receptors detect the presence of ligands on cells that are in a

“distressed” state. These include stress ligands in mouse (e.g. RAE1, H60, and MULT1) and humans (e.g. ULBP1-3 and MICA/B) that bind to the NKG2D activating receptor on

NK cells. Additionally, virally encoded non-self molecules can be expressed on the cell surfaces (e.g. CMV encoded m157) of virally infected cells and preferentially activate

NK cells. In addition to their activating receptors, NK cells also possess an array of inhibitory receptors that are used to survey cells for the absence of constitutively expressed self-ligands. For instance, MHC-I expression is recognized by the inhibitory receptor killer cell immunoglobulin-like receptors in humans, lectin-like Ly49 dimers in mice, and CD94-NKG2A heterodimer in both species 341.

Taken together, by identifying cells that have absent MHC-I expression and/or upregulation of stress or virally encoded ligands for NK activating receptors, a target cell can be susceptible to NK mediated lysis. By characterizing the NK ligand signature following OV infection, and deciphering the receptor-ligand interactions that are responsible for the differential clearance of virally infected cells over uninfected gliomas, a mechanistic understanding can potentially guide the development of therapeutic approaches to selectively target the NK receptors implicated in the anti-OV response.

49

Section 9—NK deficiencies

NK cell deficiencies, while a rare phenomenon, provide valuable information about the role of these cells in antimicrobial defense. While these deficiencies are relatively uncommon, they indicate the essential nature of these cells in host defense. The most informative group of disorders involves an isolated human NK cell deficiency that is associated with a specific gene mutation105. The only known human gene alteration resulting in isolated NK cell deficiency results from a polymorphism in CD16, the IgG Fc receptor which is activated following binding to IgG. In this polymorphism, the CD16 epitope recognized by mAb B73.1 is changed by a TA substitution at position 230 resulting in L48H 105. Individuals homozygous for this alteration have phenotypically normal NK cells but are not recognized by mAb B73.1344,345. Several individuals have been documented to have a homozygous 48H phenotype and they reported to have recurrent viral infections. In particular, a 5-year old girl was documented to have frequent respiratory infections, recurrent HSV stomatitis, and recurrent herpetic whitlow345. This child was deficient in NK cell cytotoxicity against K562 target cells but had normal antibody dependent cellular cytotoxicity (ADCC). Taken together, this 48H phenotype suggests the importance of this epitope in resistance to viral infections.

A second group of NK cell deficiencies result from unknown gene mutations. The most striking example of human NK cell deficiency is from a female adolescent with an absolute NK cell deficiency based on a lack of lymphocytes expressing CD56 or CD16 and an absence of both NK cell cytotoxicity and ADCC. This patient presented with

50 disseminated, life-threatening varicella infection and subsequently developed both CMV pneumonitis and cutaneous HSV106. There have also been reports of individuals with functional NK deficiency in which NK cells are present as a normal percentage of peripheral blood lymphocytes, but are deficient in activity. For example, four patients have been reported with widespread or invasive HSV disease, all with basal NK cell cytotoxicity against HSV-infected fibroblasts346.

Although isolated NK cell deficiencies present the opportunity to understand NK cell specific genes and NK cell roles in human antimicrobial defense, a variety of other disease have NK cell deficiency as a component. For instance, in Griscelli syndrome patients have variable immune deficiencies that typically include a marked reduction in

NK cell cytotoxicity but an ability to induce cytotoxicity upon IFN-α or IL-2 stimulation347-349. Despite this variable responsiveness, patients with this syndrome have a propensity for EBV and HSV infections347. In leukocyte adhesion deficiency (LAD), patients have elevated peripheral leukocytes and in some patients, a corresponding recurrence of HSV infection350,351. In these patients, the per lymphocyte ability of NK cells to mediate cytotoxicity, ADCC, or kill HSV infected target cells is severely attenuated. In LAD, there are variety of mutations in the β2 integrin CD18352. This results in the inappropriate expression of various key adhesion complexes including

LFA-1 (CD11a/DC18) and Mac-1 (CD11b/CD18). Notably, LFA-1 associates with the immunoglobulin-like activating receptor DNAM-1353. Taken collectively, the findings from both isolated human NK cell defects and diseases that include NK cell deficiencies

51 suggest that human NK cell activity is especially important in the defense against herpes infection.

52

Figure 2.1: Understanding and targeting the host response to oncolytic viral therapy is needed for its clinical success.

(A) The host response to OV therapy represents a unique interplay between cellular and secreted host factors that have the capacity to both limit viral efficacy and elicit enhanced tumor killing. (B) Due to the dichotomous nature of virus elicited host responses, continuing efforts are needed to clarify the contribution of components of the tumor microenvironment that both limit and enhance viral replication and spread. As the most critical factors are elucidated, they must be translated into pharmacological targets that can be paired with OV to result in additive or synergistic tumor cell killing.

53

Figure 2.2: Viral infection elicits a variety of antiviral cellular responses.

Following viral infection, viral pathogen associated molecular patterns are detected through both TLR and RIG pathways. Following the activation of each pathway, signals are relayed to interferon regulatory factors and NF-kB, leading to their translocation from the cytoplasm into the nucleus. Upon arrival in the nucleus, they activate the transcription of a variety of antiviral mediators that limit viral replication and spread.

54

Figure 2.3: The tumor microenvironment following oncolytic viral infection undergoes dynamic changes that can be targeted with pharmacological co-therapy.

(A) Viral inoculation results in the infection of neoplastic tissue; however, only cancerous cells will support active viral replication. Hours following viral inoculation, the tumor microenvironment undergoes dynamic changes (B) that create a barrier for efficient viral replication and spread: (i) an angiogenic response; (ii) elaboration of inflammatory cytokines; (iii) the recruitment of components of the innate immune system; and (iv) an extracellular environment that limits viral dissemination. With these responses in place, viral clearance will be seen within days of viral administration with limited tumor killing (C); however, each component of the host responses also provide a drug target that can be used to enhance OV efficacy using co-therapy (D). 55

Figure 2.4: Viral infection induces an NK mediated inflammatory response.

In response to viral infection, NK cells are rapidly recruited to the site of infection and adopt an activated, proliferative, and cytotoxic state. In addition, through their IFN-γ production NK cells mediate macrophage activation. Following macrophage stimulation, they adopt an inflammatory M1 phenotype while elaborating secreted factors that activate

NKs cells in a dynamic feedback loop.

56

Table 2.1: Features of oncolytic viruses being used for glioma therapy

OV Genome Structure Salient Properties

• Mutants engineered targeting p16/RB or dsDNA Enveloped MEK Herpes • High transgene capacity Low number of initial viral particles Simplex • 120-200 150-200 needed for infection and spread Virus • Drugs available to limit uncontrolled viral kb nm replication • Widespread immunity to the virus in human population Non- • Mutants designed to target p16/RB and dsDNA enveloped p53 Adenovi Icosahedral Low transgene capacity rus • • High number of initial viral particles 36-38 kb 70-90 nm needed for tumor clearance • Widespread immunity to the virus in human population Newcast Enveloped • Selective replication in cells with aberrant ssRNA interferon signaling le Helical • High viral progeny from infected cells Disease 150-300 • Due to its RNA genome, high rate of 16-20 kb mutation Virus nm • No immunity in human population Non- • Selective replication in cells transformed dsRNA enveloped with ras overexpression

Reovirus Icosahedral • High viral progeny from infected cells • Due to its RNA genome, high rate of 22-27 kb 60-80 nm mutation • Widespread immunity to the virus in human population

57

Chapter 3: The in vivo NK cell response to oHSV is detrimental to viral oncolysis

Introduction

Glioblastoma remains a formidable challenge due to its devastating effects in terms of neurologic morbidity and mortality to the patient, emotional challenges to their family, and overall cost to society8. Current standard of therapy remains largely palliative with anecdotal examples of multi-year survivorship354. Multiple current and past attempts to improve this survival have involved novel chemotherapies, innovative forms of radiation and electrical/physical treatments, signal and receptor transduction inhibitors, immunotherapies, gene therapies, and oncolytic viral therapies355-358. In virotherapy, replication-defective viral or non-viral vectors are used to transfer cytotoxic genes into tumor cells. In virotherapy with oncolytic viruses, naturally occurring or genetically engineered viral mutants that selectively replicate in tumor vs. normal cells are employed to directly lyse the neoplastic cell and/or to engender an immune response against virally infected tumor cells. One such OV is oncolytic Herpes Simplex viruses: a few different types of oHSV have been genetically modified to replicate solely in tumor cells and lyse the tumor mass through multiple rounds of viral replication. To date, oHSV injected into human glioblastomas in clinical trials have been well tolerated, but efficacy has not been as forthcoming as hoped for, even taking into account the arduousness of demonstrating this in early phase trials71,80,186.

58

As a possible explanation, we have hypothesized that the host immune response to oHSV therapy for glioblastoma is a barrier towards achieving clinical success. This hypothesis is based on animal studies showing that transient depletion of phagocytic macrophage populations or pharmacologic suppression of immunity improves the effect of various oHSV86,90,118. The initial host recognition of a foreign pathogen (e.g. oHSV) consists of a variety of humoral and cellular responses designed to eliminate the biological agent and link to adaptive immunity359. However, the hypothesis that innate immunity is deleterious to virotherapy86,90,91,117,118,195,360 runs counter to the argument that such immune responses are beneficial since they facilitate an antitumor immune response. In fact, the host immune response following oHSV administration in vivo has been shown to provoke an antitumor immune response against oHSV infected cells and bystander tumor cells187,191,299,361-368. Resolution of these apparently discordant views369,370,371,372 is significant, because one would either attempt to evade or increase immunity to improve efficacy based on which mechanism of action appears to be operative.

In this context, NK cells are the perfect foe or friend of virotherapy. NK cells are rapidly recruited to the site of viral infection and mediate viral clearance and thus could be a foe360. However, they also possess tumor—clearing properties and thus stimulation of

NK cell infiltration by oHSV could actually make them a facilitator of antitumor efficacy299,361,363,364,368. In the context of oHSV therapy, the antiviral vs. antitumor role of

59

NK cells has been undefined. Additionally, the mechanism by which NK cells eradicate virally infected cells is currently a field of intense investigation112,373.

In the data presented in Chapter 4, we show that NK cell recruitment to the site of oHSV infection of experimental glioblastoma is rapid, characterized by an activated phenotype, and coordinates an inflammatory microenvironment through microglia/macrophage activation. This response does not facilitate antitumor effects, rather it leads to premature viral clearance and limits oHSV anticancer efficacy.

Methods

Cell culture

The glioblastoma cells used include the U87dEGR human glioblastoma cell line and the murine glioblastoma cell line KR158dEGFR374. Additionally, African green monkey kidney (Vero) cells were also used. These cells were cultured in Dulbeco’s Modified

Eagle’s Medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, penicillin (100 U/ml), and streptomycin (100 µg/ml). Cells were cultured at 37oC supplemented with 5% CO2.

Animal studies

Athymic mice, C57BL/6 mice (Charles River Laboratories, Wilmington, MA), and

B6.129S7-Ifngtm1Ts/J (IFN-γ KO) (The Jackson Laboratory, Bar Harbor MA) were anesthetized by intraperitoneal administration of ketamine (100mg/kg)/xylazine

60

(20mg/kg) and stereotactically injected with glioblastoma cells into the right frontal lobe of the brain (2mm lateral and 1mm anterior to bregma at a depth of 3 mm). A total of 1 x

105 human U87dEGFR glioblastoma cells were implanted into athymic mice while 4 x105 murine KR158dEGFR glioblastoma cells were implanted into C57BL/6 or IFN-γ

KO mice. For NK depletion experiments, either 400 µl of TMβ1 or 50 µl asialo-GM1

(Wako, Tokyo, Japan) antibody combined with 50 µl water or vehicle control was injected intraperitoneally two days prior to oHSV administration. The U87dEGFR cells were allowed to grow for 9 days while the KR158dEGFR cells grew for 10 days. Unless otherwise noted, animals were then randomly divided into groups that were injected intratumorally with either rQnestin34.5 in 3µl of HBSS or vehicle. For experiments with

WT HSV1 (F Strain), 104 plaque forming units of virus was inoculated into the right frontal lobe of glioblastoma free athymic mice.

Flow cytometric analysis

Mononuclear cells were isolated from oHSV-infected brains using a previously described procedure with minor modifications375. In brief, 72 hours after infection, mice were perfused with PBS where noted, sacrificed, brain tissue was harvested, and the tumor bearing hemispheres was placed in DMEM. The tissue was homogenized through a 70

µm strainer. Single cell preparations were re-suspended in 30% Percoll, overlaid on 70%

Percoll (GE Healthcare, Uppsala, Sweden) and centrifuged at 1300 x g for 30 minutes at

4oC. Cells at the 70%-30% interface were collected, and washed with PBS. For analysis of NK levels in the spleen, the tissue was collected and homogenized through a 70 µm

61 strainer. Erythrocytes were lysed using RBC Lysis Buffer (Biolegend, San Diego, CA) and washed with PBS. Cells isolated from either the brain or the spleen were treated with

Fc Block (BD, San Jose, CA). Cells were then stained with anti-mouse immune cell surface markers for 30 min at 4oC. The following anti-mouse antibodies were used:

CD3-FITC, DX5-PE, CD3-PercP, CD62L-APC, CD11b-PE, CD27-PE, and iNOS-FITC

(BD); CD62L-FITC, CD69-FITC, NKG2D-APC, CD27-FITC, Ly49d-APC, CD94-

FITC, NKp46-FITC, CD127-FITC, CD117-FITC, NKG2A-FITC, MHCII-FITC, CD86-

PercP, CD11b-PercP, CD3-APC, CD45-APC, TNF-α-FITC, CD107a-FITC

(eBioscience, San Diego, CA); Ly-6c-FITC (Biolegend); and DX5-APC (Miltenyi

Biotec, Auburn, CA). For CD107a staining, isolated mononuclear cells were cultured in

10% RPMI with monensin (eBioscience) for 4 hours, washed with PBS, and then cell surface staining was performed. For TNF-α staining, isolated mononuclear cells were cultured in 10% RPMI with Golgi-Plug (BD) for 4 hours, washed with PBS, and then intracellular staining was performed. For intracellular staining of both TNF-α and iNOS, initial surface staining was performed for 15 minutes at 4oC. Cells were then washed, re- suspended in cytofix/cytoperm (BD) for 20 minutes at 4oC, washed again, and re- suspended in 1x Perm/Wash. Intracellular staining was then performed for 20 minutes at

4oC. Lastly, following antibody staining, cells were re-suspended in 1% formalin and analyzed using a FACS Calibur (Becton Dickinson, Mountain View, CA).

62

Viral yield assay

Titers of infectious virus particles recovered form virus inoculated brains were measured as follows. Athymic mice bearing 9 day old U87dEGFR tumors were inoculated intratumorally with 5 x 105 plaque forming units of rQnestin34.5. Seventy-two hours later, mice were sacrificed and freshly removed cerebra were placed in 2 mL DMEM supplemented with penicillin-streptomycin, manually homogenized with sequential passage through 19, 22, and 25 gauge needles, and frozen and thawed for three cycles.

After a final thaw, the lysate was sonicated for 10 seconds and the cellular debris was pelleted by centrifugation. Virus containing supernatant was collected and used for virus titration on Vero cells.

Quantitative real-time reverse transcriptase PCR

Total RNA from tumor bearing hemispheres was isolated using the RNeasy Lipid Tissue

Midi kit (Qiagen, Valencia, CA). A total of 5 µg of total RNA was reverse transcribed using random hexamers and the SuperScript First-Strand cDNA Synthesis System

(Invitrogen). Quantitative real-time PCR was done with cDNA samples diluted 1:100 in water and performed using SYBR Green PCR Master Mix and an ABI PRISM 7500 sequence detection system (Applied Biosystems, Foster City, CA). The primers used were as follows: 5’-GAACGGAGATCAAACCTGCCT-3’ and 5’-

TGTAGTCTTCCTTGAACGACGA-3’ for CXCL9; 5’-

TGGAGGAACTGGCAAAAGGA-3’ and 5’-TGTTGCTGATGGCCTGATTG-3’ for

IFN-γ; 5’-CAGCTGGGCTGTACAAACCTT-3’ and 5’-

63

CATTGGAAGTGAAGCGTTTCG-3’ for iNOS; 5’-

CATATGGAATCCAACTGGATAGATGTAAGATA-3’ and 5’-

CATATGCTCGAGGGACGTGTTGATGAACAT-3’ for IL-15; 5’

AAGCCTGTAGCCCACGTCGTA-3’ and 5’- GGCACCACTAGTTGGTTGTCTTTG-

3’ for TNF-α; 5’- TGAATCCGGAATCTAAGACCATCAA-3’ and 5’-

AGGACTAGCCATCCACTGGGTAAAG-3’ for CXCL10; 5’-

GGCTGCGACAAAGTTGAAGTGA-3’ and 5’-

TCCTGGCACAGAGTTCTTATTGGAG-3’ for CXCL11; and 5’-

AAATGGTGAAGGTCGGTGTG-3’ and 5’- TGAAGGGGTCGTTGATGG-3’ for

GAPDH internal control. For sorting brain leukocytes, non-overlapping populations of cells stained with anti-mouse CD45-APC and anti-mouse CD11b-PE were separated using a FACS (BD FACSAria). Twenty-thousand cells were used to isolate total RNA from the sorted macrophage and microglia cell populations using RNeasy Micro kit

(Invitrogen) and analyzed by quantitative real-time RT-PCR. We also used a Mouse

Inflammatory Cytokines & Receptors RT2 Profiler PCR Array (Super Array Bioscience

Corporation, Frederick, MD), according to the manufacturer’s instructions, to evaluate changes in the expression of genes encoding 84 mouse cytokines and their receptors in brain tumor tissue in response to oncolytic virus treatment (alone or in combination with

NK depletion) relative to expression in vehicle treated mice. The array includes controls to assess cDNA quality and DNA contamination.

64

Results

Oncolytic virus induces rapid natural killer cell recruitment in both human xenograft and syngeneic murine glioblastomas.

We first sought to determine whether there was an increase in NK cell infiltration after administering the oncolytic herpes simplex virus type 1, rQnestin34.5376, into orthotopic human glioblastoma (U87dEGFR) xenografts and syngeneic murine glioblastoma

(KR158dEGFR) in mice. This virus has been genetically modified to replicate primarily in glioblastoma and/or other tumor cells based on both the gene disrupting insertion of a

GFP cassette into the HSV-1 ICP6 locus encoding ribonucleotide reductase (providing selectivity for p16-/- cells)10 and the insertion of a tumor specific nestin promoter which

drives expression of its γ1 34.5 neurovirulence gene (providing selectivity for nestin- positive gliomas)376 (Figure 3.1). FACS analysis was used to quantify the presence of the pan-NK marker DX5 in explanted brain tumors 2, 6, 24 and 72 hours after treatment with the indicated virus or vehicle. A significant increase in the percentage of NK cells

(DX5+CD3-) recruited into tumor bearing hemispheres was observed as early as two hours after infection (Figure 3.2a). Compared to mice treated with vehicle or heat inactivated oHSV, the percentage and total number of NK cells recruited continued to increase up to 72 hours after oHSV infection (Figure 3.2b-c) and this response was recapitulated in an immunocompetent syngeneic model (Figure 3.3-4). The presence of

NK cells in the brain was also confirmed in oHSV treated mice that were perfused prior to mononuclear cell isolation; however, the number of NK cells in vehicle treated mice was significantly abrogated (Figure 3.3)

65

As an additional control, we also confirmed that this temporal NK cell response was replicated by intracranial administration of wild-type HSV1 (F strain) into athymic mice lacking tumor (Figure 3.3, Figure 3.5). More significantly, this NK cell recruitment following oncolytic viral infection was only observed when oHSV was injected into tumor (U87dEGFR+rQnestin34.5, 72hr) (Figure 3.6), further demonstrating that viral replication in glioblastoma was needed for NK cell infiltration. This result thus showed that oHSV elicited a rapid and significant elevation of NK cells into injected human glioblastoma xenografts and syngeneic murine glioblastoma.

oHSV inoculation induces an activated NK phenotype.

Although the above findings indicated that oHSV infection led to recruitment of NK cells into the tumor, the status and physiologic role of recruited NK cells was uncertain. We thus attempted to evaluate if they expressed markers characteristic of an activated phenotype. Surface expression of 11 different surface antigens was evaluated on NK cells by FACS-analysis of oHSV- or mock-infected human glioblastoma xenografts. We found that oHSV administration induced a unique NK phenotype for cells recruited to the site of infection. Notably, recruited NK cells exhibited significantly increased surface expression of the early activation marker CD69377, the lymphocyte homing antigen

CD62L378, and the activating receptor NKG2D379 (Figure 3.7, Table 3.1). Additionally, the activating receptor Ly49d380 and the developmental CD27340 antigen were also increased. This observed increase was independent of immunocompetence since it also

66 occurred in a syngeneic mouse glioblastoma model (Figure 3.7, Table 3.2).

Additionally, changes in peripheral NK cells following intracranial oHSV inoculation were also evaluated. Interestingly, neither the proportion nor phenotype of peripheral NK cells were altered in oHSV treated mice compared to vehicle treated mice (Figure 3.8).

These findings demonstrate that the recruited NK cells were activated locally within the tumor microenvironment and that this activation was not based on a trivial species-based recognition.

oHSV therapy induces the recruitment of distinct cytotoxic NK subsets.

NK cells can be divided into separate functional subsets according to CD27 and CD11b expression340,381: a senescent CD11bhighCD27low population, a highly proliferative and cytotoxic CD11bhighCD27high population, and an immature CD11blowCD27high population.

We thus tested the response of these subsets to oHSV infection of glioblastoma. In both athymic and syngeneic tumor models, the majority of NK cells recruited to vehicle treated mice were CD11bhighCD27low. In contrast, oHSV administration led to the recruitment of either CD11bhighCD27high or CD11blowCD27high NK cells (Figure 3.9a-b,

Table 3.3). As an additional control, CD11b/CD27 expression was equivalent when comparing tumor-free mice inoculated with wild type HSV and tumor bearing mice treated with oHSV (Figure 3.9a, 3.10). Lastly, we were interested in comparing the

CD11b/CD27 profile of NK cells circulating within the tumor-bearing hemisphere early during tumor formation (3 days post-U87dEGFR implantation) and after treating a 9 day

67 old tumor with vehicle (U87dEGFR+Veh.). These experiments revealed that the NK profile was consistent during early and later tumor formation (Fig 3.10).

We also evaluated NK cytotoxicity by examining expression of the degranulation marker

CD107a. Within the CD11bhighCD27low population, oHSV administration induced a 3- fold increase in CD107a staining compared to vehicle treated mice, while in the

CD11bhighCD27high subset oHSV inoculation resulted in a 7-fold increase in CD107a staining compared to control treated mice (Figure 3.9c). Taken together, these findings demonstrate that oHSV therapy significantly enhances the recruitment of distinct NK subsets expressing proliferative/cytotoxic NK cell markers.

Macrophages and microglia are activated following oHSV therapy in a NK dependent manner.

NK cells are thought to not only provide innate immune functions against tumors and viruses, but also to coordinate additional innate cellular responses, such as activation of mononuclear and macrophage cell populations. We thus sought to determine if the recruited NK cells activated by oHSV infection of glioblastoma, were also able to further coordinate macrophage recruitment and activation382. Using FACS, oHSV treatment resulted in a relative decrease in the percentage of endogenous microglia and a concomitant increase in macrophage and lymphocyte populations compared to vehicle treated mice (Figure 3.11). These mononuclear cells also displayed significantly enhanced surface expression of macrophage activation markers MHC-II, Ly-6C, and

68

CD86 (Figure 3.12). To test the role of NK cells in mediating the observed recruitment and activation of macrophages, we found that NK depletion with either Asialo-GM1 or

TMβ1 (commonly utilized methods for NK cell depletion383-386) significantly attenuated the cell surface staining of canonical macrophage activation markers, without significantly altering the proportion of macrophages recruited into the brain following viral infection (Figure 3.13). As expected, the inset of Figure 3.13 shows that either depletion method significantly reduced the lymphocyte population, including NK cells, but not the macrophage population recruited to the infected brain tumor.

We then proceeded to analyze both transcript and protein levels for two markers of macrophage/microglia activation: iNOS and TNF-α. While oHSV infection of glioblastoma significantly increased their gene expression in tumor hemispheres, NK depletion abrogated this response (Figure 3.14a). We were interested if this NK coordinated response was a product of altered NK mediated recruitment of macrophage and microglia cells to the site of infection or a NK mediated change in their activation status. Using FACS and RT-qPCR to assess gene expression from an equal number of macrophage and microglia cells, we found that increases in iNOS and TNF-α gene expression following oHSV inoculation was a product of NK mediated activation of these cells rather than altered recruitment of macrophages and microglia to the site of infection.

(Figure 3.14b). Confirming our gene expression data, we detected increased iNOS and

TNF-α FACS staining in the macrophage and microglia compartments following oHSV administration, and NK depletion was able to reverse this response (Figure 3.15).

69

Additionally, we confirmed that iNOS induction at the gene and protein levels was mediated in an IFN-γ dependent manner (Figure 3.16).

Finally, we analyzed the activation status of macrophages and microglia by studying the expression level of the IFN-γ inducible chemokines (IIC) CXCL9, CXCL10, and

CXCL11. Consistent with our NK infiltration and macrophage/microglia activation findings, we found that IIC expression is robustly increased following oHSV therapy

(Figure 3.17a). However, when NK cells were depleted or an IFN-γ knockout (KO) mouse was used, the level of these chemokines was significantly attenuated (Figure

3.17b). Consistent with the notion that IIC expression is associated with microglia/macrophage activation but is not necessarily derived from these cells, the expression of these chemokines was not significantly altered within the macrophage and microglia populations following NK depletion (Figure 3.18). Taken collectively, these findings provide evidence that recruited and activated NK cells by oHSV-infection of glioblastoma were critical in the coordination of microglia/macrophage activation which has also been reported to correlate with anticancer and antiviral properties387,388.

Depletion of NK cells in vivo leads to enhanced viral replication and improves glioblastoma therapy.

Since NK cells possess both anti-tumor and anti-viral properties, we did not know if the observed recruitment of an activated NK cell subpopulation and its subsequent coordination of activating microglia and macrophages would lead to improved

70 glioblastoma therapy by eliminating tumor cells or to reduced efficacy by eliminating oHSV and oHSV-infected glioblastoma cells. To test this we compared survival of glioblastoma bearing mice treated with oHSV in the presence or absence of NK depleting antibody. We first confirmed our ability to deplete NK cells (Figure 3.19).

Glioblastoma bearing mice treated with both oHSV and NK depletion had significantly elevated viral titers compared to identical mice not treated with NK depletion (Figure

3.20a). While NK depletion did not impact overall survival of glioblastoma bearing mice treated with HBSS, there was a significant prolongation in the median survival of glioblastoma bearing mice treated with oHSV and NK depletion, regardless of the type of

NK depletion (anti-asialo-GM1 or anti-TM-β1; p < 0.05) (Figure 3.20b).

We used a gene expression array to test changes in 84 different mouse inflammatory genes in mice treated with oHSV in the presence or absence of NK cell depletion (Figure

3.21, Table 3.4-5). Consistent with previous findings91, we found a significant induction in 30 of the 84 genes analyzed in tumors treated with oHSV compared to vehicle treatment, with over 100-fold induction of the chemokines CXCL10 (588-fold), CXCL9

(232-fold), CXCL11 (139-fold), CCL7 (117-fold), and CCL2 (104-fold). Interestingly, thirteen of these genes were markedly reduced in tumors obtained from mice treated with oHSV after NK cell depletion. Collectively, these results indicate that NK cell mediated inflammation impedes viral propagation and overall survival of oHSV treated mice and are not supportive of a model where activation of NK cells leads to improved anticancer.

71

Discussion

Oncolytic virotherapy and oHSV therapy of glioblastoma continues to be tested in several clinical trials around the world. It has so far been well tolerated and evidence of efficacy will require more advanced phase trials. It remains unclear whether the mechanism of

OV antitumor efficacy depends on immune-mediated mechanisms leading to rejection of tumor and virally-infected tumor cells or on a direct lytic effect by OV. This is of critical significance since knowledge of this mechanism would translate into pursuing opposite approaches to OV therapy, namely immunostimulatory vs. immunosuppressive. In this study, we focused on the role of innate immunity, specifically the NK response to oHSV therapy of glioblastoma, and whether it helped or hindered antitumor efficacy and overall mouse survival. We discovered a rapid NK cell response that specifically results from tumor infection with oHSV. We show that this in vivo antiviral NK cell response to oHSV-infected glioblastoma is unique to the brain, is dependent on the activation of specific cytotoxic NK cell subsets that also coordinate the activation of macrophages, and occurs in both human xenograft and mouse syngeneic models of glioblastoma.

Importantly, we have discovered that this NK cell response leads to clearance of oHSV- infected glioblastomas and that down-regulation of this response by depletion of NK cells actually leads to improved survival of mice with glioblastoma treated with oHSV. This study thus demonstrates that the observed rapid host NK cell response to glioblastoma following oHSV therapy functions to cooperate with other innate immune effector populations to reduce, rather than enhance, the efficacy of oHSV treatment.

72

Before this current study, the role of NK cells and innate immunity in mediating the efficacy of virotherapy was at best controversial. In fact, contrary to our findings, a majority of investigators have shown that oHSV efficacy was actually enhanced by eliciting an NK mediated anti-tumor response299,361,363,364. Only Altomonte et al. showed that the antiviral properties of NK cells are detrimental to vesicular stomatitis virus therapy for hepatocellular carcinoma389. Based on our initial evidence that oHSV elicits a robust NK response and our previous findings that CPA modulation of innate immunity enhances oHSV mediated glioblastoma killing90,91,117,118,195,390, we used NK depletion to confirm their antiviral role in the context of oHSV treatment of glioblastoma.

Additional evidence supports the hypothesis that the rapid response of NK cells to oHSV infection of glioblastoma provides the initial line of host defense against the virus and is actually deleterious to efficacy. First of all, there is a decline in viral titers that occurs within days of inoculating various oncolytic viruses90. Second, preclinical findings have found that the clearance of over 80% of oHSVs corresponds with the rapid recruitment of peripheral macrophages into the site of viral infection, suggesting that this response is responsible for mediating oHSV clearance359. These findings were validated through a study demonstrating that macrophage depletion enhanced the efficacy of oHSV therapy in glioblastoma86. Third, a recently published clinical trial noted a prominent inflammatory infiltrate within the tumor microenvironment of GBM patients treated with oHSV71, with some evidence of less than robust oHSV replication. Fourth, there have

73 been few instances of humans with congenital NK cell deficiencies that have succumbed to HSV1 encephalitis104,106. Lastly, mathematical modeling has previously shown the timing of the antiviral innate immune response is detrimental to oHSV therapy195. Taken in combination, this evidence appears to strongly suggest that initial innate immunity to oHSV therapy is detrimental and deleterious to anticancer efficacy, rather than helpful.

However, other studies have shown that immune responses in the context of virotherapy based on vesicular stomatitis virus (VSV) for metastatic non-CNS tumors, for example, are highly desirable363,180,391,392 and lead to effective antitumor immunity. Therefore, the effect of immunity on efficacy of virotherapy may require a balance between oncolytic virus replication and antiviral/antitumor responses that may differ based on the OV used and the tumor type and model.

Although oHSV infection has been previously reported by us to result in the recruitment of NK cells, the functional relevance of these cells in vivo was unclear202. For the first time, we show that these recruited murine NK cells up-regulate markers of an activated phenotype (e.g. CD11bhighCD27high expressing the CD107a degranulation marker).

Although we have previously demonstrated that macrophages and microglia are a limiting factor in the successful oHSV treatment for glioblastoma, the role of NK cells in mediating this response had not been studied. This is particularly relevant since one of the main products of NK cells, IFN-γ, is a prototypical macrophage activator. With the knowledge that oHSV therapy is enhanced in both IFN-γ and macrophage depleted mice86,90, we hypothesized that NK cells are likely mediators of oHSV-induced

74 macrophage activation based on a variety of markers that are associated with HSV induced inflammation. We demonstrated that macrophage/microglia activation occurring following oHSV administration is mediated in a NK cell and IFN-γ dependent manner.

The results of figure 3.14b argue that the loss of macrophage activation products were more likely due to loss of NK-mediated macrophage activation, rather than to loss of macrophage themselves.

Interestingly, oHSV infection led to a robust NK cell and IFN-γ dependent induction of

CXCL9, CXCL10, and CXCL11. These chemokines are associated with macrophage activation and the induction of a potent antiviral inflammatory response leading to herpes encephalitis396. Translating these findings into humans may mean that recruitment of

CD3-CD56brightCD94bright human NK cells along with their by product, IFN-γ, could be associated with oHSV inoculation, with the potential of mediating a deleterious antiviral inflammatory response to oHSV through macrophage and microglia activation.

Similarly, the CD3-CD56dimCD94dim subset that is associated with cell-mediated killing may translate into preferentially killing virally infected glioblastoma cells before sufficient oncolysis has occurred. Thus, future human clinical trials studies could clarify developmental NK subsets, by examining CD56 and CD94 expression397, along with their associated functional properties.

One potential limitation of our study may be that we have not assessed the effect of NK depletion in a fully immunocompetent mouse model. In general, replication of HSV1 is

75 not very robust in murine glioblastomas compared to human glioblastomas. Over the years, we have screened several murine glioblastomas and have found one (the murine

KR158dEGFR glioblastoma line) where oncolytic HSV replication and cytotoxicity approximates that observed with human glioblastomas. However, these glioblastomas are syngeneic to C57BL/6 mice, a strain that is notoriously impervious to HSV1, due to its expression of an HSV resistance genomic locus that may encode antiviral factor(s)398,399. In fact, attempts to use rQnestin34.5 as an in vivo therapy against

KR158dEGFR glioblastomas in C57BL/6 mice brains were not successful, due to minimal in vivo viral replication (Figure 3.22). Therefore, we have been unable to assess the role of NK cell depletion in an immunocompetent model, due to the context of lack of virotherapy in this model.

To our knowledge, the finding that activated NK cells are both rapidly recruited to the site of oHSV infection and limit viral propagation is novel and clarifies the role of NK cells within the broader host-response to oncolytic HSV. Moreover, by demonstrating the critical role of NK cells in mediating the downstream activation of macrophages and microglia, we have demonstrated that NK cells are the critical orchestrators of the innate immune response to oHSV. Future studies will be needed both to extend these findings to humans and to modulate the NK response in a precise, time-dependent fashion, in order to realize the benefit of an initial reduced antiviral response while still allowing for the benefit of downstream anti-tumor immunity.

76

Figure 3.1: Construction of an oHSV that replicates specifically in tumor cells expressing nestin376.

A schematic map of wild-type HSV (F strain), rHsvQ1 (a double UL39 and γ34.5 mutant), and rQnetin34.5 is presented. In the middle construct, green fluorescent protein

(GFP) cDNA is inserted into the UL39 locus and both γ34.5 genes have been deleted.

The bottom construct demonstrates the site of recombination of the hybrid promoter

(nestin enhancer and hsp68 minimum promoter) combined with a γ34.5 expression cassette into ICP6, giving rise to the mutant oHSV rQnestin34.5.

77

A. continued

Figure 3.2: NK cells recruitment to the site of oHSV infection increases in a time dependent manner.

(A) Human U87dEGFR cells (105) were implanted intracranially into athymic mice brains (n=3/group) and allowed to grow for 9 days. rQnestin34.5 (104 pfu/3µl vehicle) or vehicle was then stereotactically inoculated, using the same coordinates. Tumor bearing hemispheres were harvested 2 hours later to quantify the percentage of NK cells with

FACS. (B) Athymic mice were implanted with tumor and treated as described in (A).

Tumor bearing hemispheres were harvested 6, 24, or 72 hours after infection (n=4-

5/group) so that the total number of NK cells in tumor bearing hemispheres could be quantified using FACS. (C) Athymic mice were implanted with tumor, treated, and sacrificed as described in (B). The percentage of infiltrating NK cells in tumor bearing hemispheres could be quantified using FACS. * P < 0.05; ** P < 0.01; *** P < 0.001.

Error bars represent +/- standard deviation. 78

Figure 3.2 continued

B.

C.

79

Figure 3.3: The infiltrative NK response occurs in both xenograft and syngeneic glioblastoma models.

Representative dotplots of CD3-DX5+ NK cells are presented from various animal models 72 hours after viral inoculation. These include athymic nude mice implanted with

105 U87dEGFR tumor and then treated with either 104 pfu rQnestin34.5 or wild type

HSV-1 (F Strain); or C57BL/6 mice implanted with KR158dEGFR murine glioblastoma

(4x105 cells) followed by rQnestin34.5 (105 pfu/3µl) inoculation 10 days after tumor implantation. In dotplots noted as “Perfused”, mice were perfused with PBS prior to processing the brain for NK isolation. The indicated percentages are CD3-DX5+ NK cells.

80

Figure 3.4: The NK response to oHSV is recapitulated in a syngeneic model.

C57BL/6 mice (n=4/group) were implanted with KR158dEGFR (4x105 cells) and then inoculated with 105 pfu rQnestin34.5 ten days after tumor implantation. Mice were sacrificed 72 hours later and the total number of NK cells was assessed by FACS. *** P

< 0.001. Error bars represent +/- standard deviation.

81

Figure 3.5: Oncolytic and wild-type viral infection results in a similar temporal NK response.

Human U87dEGFR cells (105) were implanted intracranially into athymic mice brains

(n=3/group) and allowed to grow for 9 days. rQnestin34.5 (104 pfu/3µl vehicle), F-strain wild-type HSV-1 (104 pfu/3µl vehicle), or vehicle was then stereotactically inoculated, using the same coordinates. Tumor bearing hemispheres were harvested 72 hours later to quantify the percentage of NK cells with FACS.

82

Figure 3.6: Active viral replication is required for NK cell recruitment.

The total number of NK cells was quantified by FACS for the following conditions: 72 hours after intracranial injection of 104 pfu rQnestin34.5 into 9 day old tumor

(U87dEGFR+rQnestin34.5); 72 hours after intracranial injection of 104 pfu of wild type

HSV (F-strain) into tumor free hemisphere; 12 days after U87dEGFR tumor implantation; 72 hours after intracranial injection of vehicle into 9 day old tumor

(U87dEGFR+Veh.); or 72 hours after intracranial injection of 104 pfu of rQnestin34.5 into tumor free hemisphere. *** P < 0.001. Error bars represent +/- standard deviation.

83

Figure 3.7: NK cells are activated following oHSV therapy.

Glioblastoma free athymic mice, athymic mice with 9 day old U87dEGFR human glioblastoma (105 cells originally implanted), or immunocompetent C57BL/6 mice with

10 day old KR158dEGR murine glioblastoma (4 x 105 cell originally implanted) were treated with vehicle or rQnestin34.5 (for U87dEGFR or KR158dEGFR) or Wild type

(WT) HSV for glioblastoma free brain (n=4-6 mice/group). rQnestin34.5 or WT HSV

4 5 (10 pfu/3µl) was used for athymic mice while C57BL/6 mice were treated with 10 pfu/3µl rQnestin34.5. Seventy-two hours later, mice were sacrificed and tumor bearing hemispheres were processed so that the NK cells could be analyzed using FACS.

84

Figure 3.8: Peripheral NK cells do not exhibit phenotypic changes following oHSV therapy.

Athymic mice with 9 day old U87dEGFR human glioblastoma (105 cells originally implanted) were treated with vehicle or 104 pfu rQnestin34.5 (n=4 mice/group). Seventy- two hours later, mice were sacrificed and spleens were processed so that the NK cells could be analyzed using FACS. Each panel is a representative dotplot to demonstrate the altered surface expression of five NK surface antigens following viral administration. 85

A. continued

Figure 3.9: oHSV administration results in the recruitment of distinct NK subsets.

(A) Glioblastoma free athymic mice, athymic mice with 9 day old U87dEGFR human glioblastoma (105 cells originally implanted), or immunocompetent C57BL/6 mice with

10 day old KR158dEGR murine glioblastoma (4 x 105 cell originally implanted) were treated with vehicle or rQnestin34.5 (for U87dEGFR or KR158dEGFR) or Wild type

(WT) HSV for glioblastoma free brain (n=4-6 mice/group). rQnestin34.5 or WT HSV

4 5 (10 pfu/3µl) was used for athymic mice while C57BL/6 mice were treated with 10 pfu/3µl rQnestin34.5. Seventy-two hours later, mice were sacrificed and tumor bearing hemispheres were processed so that the NK cells could be analyzed using FACS for

CD11b/CD27 co-expression on CD3-DX5+ NK cells. (B) Athymic mice were treated as described in (a). The total number of tumor derived NK cells expressing CD11b/CD27 was quantified and is presented as a fold increase in oHSV treated mice versus vehicle treated mice (n=4-6 mice/group). (C) Athymic mice (n=3/group) were treated as described in (a) and the percentage positivity of CD107a in each CD11b/CD27 subset was quantified. ** P < 0.01; *** P < 0.001. Error bars represent +/- standard deviation.

86

Figure 3.9 continued

B.

C.

87

Figure 3.10: Viral infection and glioblastoma formation result in the recruitment of differential NK subsets.

Glioblastoma free athymic mice or athymic mice with 9 day old U87dEGFR human glioblastoma (105 cells originally implanted), were treated with vehicle or rQnestin34.5 or Wild type (WT) HSV (104 pfu/3µl) for glioblastoma free brain (n=4-6 mice/group).

Seventy-two hours later, mice were sacrificed and tumor bearing hemispheres were processed so that the NK cells could be analyzed using FACS for CD11b/CD27 co- expression on CD3-DX5+ NK cells. Additionally, athymic nude mice were implanted with 105 U87dEGFR cells and sacrificed 72 hours later to analyze CD11b/CD27 co- expression. ** P < 0.01; *** P < 0.001. Error bars represent +/- standard deviation.

88

Figure 3.11: Macrophages and lymphocytes recruited to the site of oHSV infection increases in a time dependent manner.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice and allowed to grow for 9 days. rQnestin34.5 (104 pfu/3µl) or vehicle was then inoculated at the site of tumor implantation and mice were sacrificed at different time points in order to quantify by FACS the percentage microglia (CD45lowCD11b+), macrophage (CD45highCD11b+), and lymphocyte (CD45highCD11blow) cell populations in tumor bearing hemispheres following treatment (n=4-6 mice/group).

89

Figure 3.12: Macrophages are activated following oHSV therapy.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice

(n=4/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb (50µl antibody+50µl

4 H2O). Two days later, 10 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later so that the percentage of macrophages expressing activation markers (MHC-II, Ly-6C, and CD86) could be assessed by FACS. * P < 0.05; *** P < 0.001. Error bars represent +/- standard deviation.

90

Figure 3.13: NK depletion does not alter the proportion of macrophages following oHSV administration.

C57BL/6 mice or IFN-γ KO mice were implanted with KR158dEGFR (4x105 cells) and allowed to grow for 8 days. Mice were then depleted of NK cells with a single intraperitoneal injection of either anti-Asialo-GM1 mAb (50µl antibody+50µl H2O) or anti-TMβ1 mAb (400µl). Two days later, 105 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later. Representative dotplots of CD45/CD11b (R1=microglia, R2=macrophage, R3=lymphocyte) staining are presented.

91

A. continued

Figure 3.14: NK cells mediate macrophage and microglia activation following oHSV inoculation.

(A) U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice (n=4-

5/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with anti-Asialo-GM1 mAb. Two days later, 104 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation, mice were sacrificed 72 hours for mRNA isolation and cDNA synthesis, and the expression of iNOS or TNF-α was evaluated. (B) Athymic mice

(n=5/group) were implanted with tumor, depleted of NK cells, and treated with rQnestin34.5 as described in (A). Mice were sacrificed 72 later and the microglia and macrophage populations were sorted using FACS. mRNA was isolated from either

20,000 microglia or macrophage cells and iNOS or TNF-α gene expression was evaluated with RT-qPCR. ** P < 0.01; *** P < 0.001. Error bars represent +/- standard deviation.

92

Figure 3.14 continued

B.

93

Figure 3.15: NK cells mediate TNF-α production from microglia and macrophages.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice

(n=4/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb (50µl antibody+50µl

4 H2O). Two days later, 10 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later so that TNF-α expression could be assessed with FACS within the microglia, macrophage, and lymphocyte populations. The percentages represent TNF-α surface staining in each cell population.

94

A. continued

Figure 3.16: iNOS production following oHSV infection is induced in an IFN-γ dependent manner.

C57BL/6 mice or IFN-γ KO mice were implanted with KR158dEGFR and allowed to grow for 8 days. Mice were then depleted of NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb. Two days later, 105 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later so that iNOS expression could be assessed with FACS within the microglia and macrophage populations. The percentages represent iNOS surface staining in each cell population.

(B) C57BL/6 mice or IFN-γ KO mice (n=3-4/group) were implanted with tumor, depleted of NK cells, and inoculated with virus as described in (A). Mice were sacrificed 72 hours later for mRNA isolation of the tumor bearing hemisphere. Following mRNA conversion to cDNA, the expression of iNOS was evaluated. * P < 0.05; *** P < 0.001. Error bars represent +/- standard deviation.

Figure 3.16 continued 95

B.

A. continued

96

Figure 3.17: NK cells and IFN-γ mediate CXCL9, 10, 11 expression.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice

(n=4-5/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with anti-Asialo-GM1 mAb. Two days later, 104 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later for mRNA isolation of the tumor bearing hemisphere. Following mRNA conversion to cDNA, the expression of CXCL9, CXCL10, and CXCL11 was evaluated. (B) C57BL/6 mice or IFN-γ KO mice were implanted with KR158dEGFR (4x105 cells) and allowed to grow for 8 days. Mice were then depleted of NK cells with anti-Asialo-GM1 mAb. Two days later, 105 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later for mRNA isolation of the tumor bearing hemisphere. Following mRNA conversion to cDNA, the expression of CXCL9,

CXCL10, and CXCL11 was evaluated. * P < 0.05; ** P < 0.01; *** P < 0.001. Error bars represent +/- standard deviation.

Figure 3.17 continued

97

B.

98

Figure 3.18: NK depletion does not alter microglia or macrophage production of IFN-γ inducible chemokine.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice and allowed to grow for 7 days. Mice were then depleted of their NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb (50µl antibody+50µl H2O). Two days later, 104 pfu rQnestin34.5 or vehicle was inoculated at the site of tumor implantation, mice were sacrificed 72 hours later, and the microglia and macrophage populations were sorted using FACS. mRNA was isolated from either 20,000 microglia or macrophage cells and CXCL10 gene expression was evaluated with RT-qPCR.

99

A. continued

Figure 3.19: Asialo-GM1 treatment depletes NK cells.

(A) Representative dotplots are shown from athymic mice that received an intraperitoneal injection of asialo-GM1 mAb or vehicle, sacrificed three days later, and the percentage of

NK cells from the spleen was quantified using FACS. Additionally, human U87dEGFR glioblastoma (105 cells) were implanted intracranially into athymic mice and allowed to grow for 7 days. Mice were then depleted of their NK cells with anti-Asialo-GM1 mAb.

Two days later, rQnestin34.5 (104 pfu/3µl) or vehicle was inoculated at the site of tumor implantation, mice were sacrificed 72 hours later, and the percentage of NK cells from the brain was quantified using FACS. (B) Athymic mice were implanted with tumor, treated with anti-Asialo-GM1 mAb, inoculated with virus, and sacrificed as described in

(A). The total number of NK cells from vehicle and rQnestin34.5 treated mice is presented (n=4 mice/group). (C) Athymic mice were implanted with tumor, treated with anti-Asialo-GM1 mAb, and inoculated with virus as described in (A). Mice were sacrificed 72 hours later for mRNA isolation of the tumor bearing hemisphere.

Following mRNA conversion to cDNA, the expression of IFN-γ was evaluated (n=4-5 mice/group). ** P < 0.01; *** P < 0.001. Error bars represent +/- standard deviation. 100

Figure 3.19 continued

B.

C.

101

A. continued

Figure 3.20: NK depletion enhances oHSV efficacy.

(A) U87dEGFR glioblastoma (105 cells) were implanted intracranially into athymic mice

(n=4-5/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb (50µl antibody+50µl

5 H2O). Two days later, rQnestin34.5 (5 x 10 pfu/3µl) or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later so that viral titers could be assessed. (B) Athymic mice (n=5/group) were implanted with tumor as described in (a).

Seven days after tumor implantation, mice were then depleted of their NK cells with an intraperitoneal injection of either anti-Asialo-GM1 mAb (50µl antibody+50µl H2O) or anti-TMβ1 mAb (400µl) with subsequent injections every 7 days. Two days later, 104 pfu rQnestin34.5 or vehicle was then inoculated at the site of tumor implantation and mice were observed for the onset of neurological symptoms.

102

Figure 3.20 continued

B.

103

Figure 3.21: NK cells mediate robust inflammatory response following oHSV infection.

U87dEGFR glioblastoma (105 cells) were implanted intracranially into athymic mice

(n=4-5/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb (50µl antibody+50µl

5 H2O). Two days later, rQnestin34.5 (5 x 10 pfu/3µl) or vehicle was inoculated at the site of tumor implantation and mice were sacrificed 72 hours later for mRNA isolation of the tumor bearing hemisphere. Following mRNA conversion to cDNA, the expression of 84 mouse inflammatory genes were analyzed by RT-qPCR and presented as a fold- difference. Each row and column is represented by a unique number denoting a different transcript. On the left, the comparison is vehicle vs rQnestin34.5 treatment while for the right, the comparison is Asialo-GM1+rQnestin34.5 vs. rQnestin34.5 treatment.

104

Figure 3.22: A lack of HSV infectability in C57BL/6 mice prevents assessment of oHSV efficacy in the KR158dEGFR glioblastoma model.

KR158dEGFR glioblastoma (4x105 cells) were implanted intracranially into C57BL/6 mice and allowed to grow for 8 days. Mice were then depleted of their NK cells with an intraperitoneal injection of either anti-Asialo-GM1 mAb (50µl antibody+50µl H2O) or anti-TMβ1 mAb (400µl). Two days later, 105 pfu rQnestin34.5 or vehicle was then inoculated at the site of tumor implantation and mice were observed for the onset of neurological symptoms.

105

Table 3.1: Phenotype of NK cells recruited to the site of infection in glioma xenograft

U87dEGFR bearing mice

% Positive MFI

Veh. rQnestin34.5 p-value Veh. rQnestin34.5 p-value

CD69 24 (4.6) 52.6 (2.1) 0.0006 22 (1.6) 32 (1.0) 0.0007

CD62L 47.2 (8.8) 57.8 (2.7) 0.028 140 (7.0) 197 (15.0) 0.008

NKG2D 85.6 (2.3) 90.2 (1.7) 0.004 24 (3.1) 34 (3.1) 0.002

CD27 31.2 (5.5) 54 (7.8) < 0.0001 21 (0.3) 25 (1.0) 0.8

Ly49d 29.2 (11.8) 21.5 (7.9) 0.19 16.3 (2.0) 23.3 (1.5) 0.01

CD11b 81.2 (4.3) 77.2 (2.6) 0.16 53 (4.3) 49 (5.0) 0.3

CD94 53 (10) 50 (4.6) 0.6 16.2 (5.5) 15.6 (5.0) 0.8

NKp46 89.2 (3.4) 92.2 (2.1) 0.18 32.5 (0.7) 31 (1.4) 0.3

CD127 16 (3.0) 23 (4.3) 0.049 10.2 (1.1) 10 (0.2) 0.8

CD117 12.7 (3.3) 12 (1.2) 0.3 16.4 (3.5) 12.7 (3.3) 0.12

NKG2A 56.6 (7.1) 54.3 (3.2) 0.63 14.2 (0.8) 13.2 (0.9) 0.21

106

Table 3.2: Phenotype of NK cells recruited to the site of infection in syngeneic glioma model

KR158dEGFR bearing mice

% Positive MFI

rQnestin34.

Veh. rQnestin34.5 p-value Veh. 5 p-value

CD69 24.3 (13.5) 55.5 (13.0) 0.046 21 (4.7) 27 (2.2) 0.03

CD62L 64 (10.2) 35.7 (5.7) 0.0009 246 (5) 270 (6.5) 0.006

NKG2D 78.8 (14.6) 63.4 (1.7) 0.047 32 (7.5) 49 (10.5) 0.01

CD27 60.5 (5.5) 30.7 (7.8) 0.0001 20 (4.9) 24 (2.0) 0.7

Ly49d 29 (12.6) 29.7 (7.6) 0.92 64 (12.9) 117 (22.2) 0.006

CD11b 75.5 (2.1) 87.6 (2.1) 0.007 42 (3.0) 41 (2.9( 0.9

CD94 40.5 (6.3) 66.5 (2.1) 0.031 35 (2.8) 32 (1.4) 0.3

NKp46 62.3 (6.6) 65.6 (4.1) 0.42 10.7 (0.4) 12 (0.5) 0.03

CD127 13 (3.5) 11.3 (1.8) 0.44 13.6 (1.6) 13.8 (0.8) 0.85

CD117 12 (9.1) 29.5 (0.7) 0.26 12.1 (3.3) 15.7 (1.4) 0.48

NKG2A 6 (1.4) 8 (1.4) 0.29 7.8 (0.3) 8.5 (0.2) 0.13

107

Table 3.3: Developmental status of recruited NK cells following oHSV infection

% % %

high low high high low high

CD11b CD27 CD11b CD27 CD11b CD27

Veh. 66.5 (6.2) 15.2 (4.6) 17.5 (5)

rQnestin34.5 43 (7.2) 27.6 (3.2) 28.3 (7.3) U87dEGFR

p-value < 0.0001 < 0.0001 0.0063

% % %

high low high high low high CD11b CD27 CD11b CD27 CD11b CD27

Veh. 66 (9.7) 19 (7.0) 12 (6.1)

rQnestin34.5 36 (4.5) 32 (2.6) 29 (3.9) KR158dEGFR p-value < 0.0001 0.0014 0.00023

108

Table 3.4: Changes in inflammatory cytokine profile between vehicle and rQnestin34.5 treated mice

Gene Fold Change

Position Description Symbol (rQnestin34.5/Vehicle) p-value

ATP-binding cassette, sub-family

A01 F (GCN20), member 1 Abcf1 -1.11 0.134586

A02 B-cell leukemia/lymphoma 6 Bcl6 -1.13 0.089836

Chemokine (C-X-C motif)

A03 receptor 5 Cxcr5 1.20 0.720905

A04 Complement component 3 C3 9.84 0.001047

A05 Caspase 1 Casp1 2.28 0.002213

A06 Chemokine (C-C motif) ligand 1 Ccl1 5.27 0.084505

A07 Chemokine (C-C motif) ligand 11 Ccl11 5.82 0.005606

A08 Chemokine (C-C motif) ligand 12 Ccl12 29.99 0.000851

A09 Chemokine (C-C motif) ligand 17 Ccl17 -1.21 0.218366

A10 Chemokine (C-C motif) ligand 19 Ccl19 1.05 0.992575

A11 Chemokine (C-C motif) ligand 2 Ccl2 104.10 0.003958

A12 Chemokine (C-C motif) ligand 20 Ccl20 2.05 0.027173

B01 Chemokine (C-C motif) ligand 22 Ccl22 2.14 0.097439

B02 Chemokine (C-C motif) ligand 24 Ccl24 -1.18 0.333739

B03 Chemokine (C-C motif) ligand 25 Ccl25 1.02 0.774018

B04 Chemokine (C-C motif) ligand 3 Ccl3 1.74 0.350660

B05 Chemokine (C-C motif) ligand 4 Ccl4 7.29 0.009293

109 continued

Table 3.4 continued

B06 Chemokine (C-C motif) ligand 5 Ccl5 23.19 0.002831

B07 Chemokine (C-C motif) ligand 6 Ccl6 1.49 0.365062

B08 Chemokine (C-C motif) ligand 7 Ccl7 117.89 0.007497

B09 Chemokine (C-C motif) ligand 8 Ccl8 52.74 0.000488

B10 Chemokine (C-C motif) ligand 9 Ccl9 1.17 0.394502

B11 Chemokine (C-C motif) receptor 1 Ccr1 2.43 0.013631

B12 Chemokine (C-C motif) receptor 2 Ccr2 7.23 0.000017

C01 Chemokine (C-C motif) receptor 3 Ccr3 2.47 0.001042

C02 Chemokine (C-C motif) receptor 4 Ccr4 1.38 0.279014

C03 Chemokine (C-C motif) receptor 5 Ccr5 2.30 0.000110

C04 Chemokine (C-C motif) receptor 6 Ccr6 -2.70 0.139646

C05 Chemokine (C-C motif) receptor 7 Ccr7 6.86 0.002738

C06 Chemokine (C-C motif) receptor 8 Ccr8 1.05 0.544158

C07 Chemokine (C-C motif) receptor 9 Ccr9 1.99 0.005003

C-reactive protein, pentraxin-

C08 related Crp 1.17 0.272019

Chemokine (C-X3-C motif) ligand

C09 1 Cxc3cl1 -1.02 0.691689

Chemokine (C-X-C motif) ligand

C10 1 Cxcl1 6.35 0.048337

Chemokine (C-X-C motif) ligand

C11 10 Cxcl10 588.68 0.004959

continued 110

Table 3.4 continued

Chemokine (C-X-C motif) ligand

C12 11 Cxcl11 139.14 0.002244

Chemokine (C-X-C motif) ligand

D01 12 Cxcl12 -1.79 0.000272

Chemokine (C-X-C motif) ligand

D02 13 Cxcl13 18.13 0.001079

Chemokine (C-X-C motif) ligand

D03 15 Cxcl15 1.04 0.575347

D04 Platelet factor 4 Pf4 -1.31 0.105358

Chemokine (C-X-C motif) ligand

D05 5 Cxcl5 1.88 0.212520

Chemokine (C-X-C motif) ligand

D06 9 Cxcl9 232.78 0.001373

Chemokine (C-X-C motif)

D07 receptor 3 Cxcr3 2.26 0.067502

Chemokine (C-C motif) receptor

D08 10 Ccr10 -1.15 0.553177

D09 Interferon gamma Ifng 43.97 0.008618

D10 Interleukin 10 Il10 1.46 0.865702

D11 Interleukin 10 receptor, alpha Il10ra 2.26 0.004415

D12 Interleukin 10 receptor, beta Il10rb 1.18 0.320365

E01 Interleukin 11 Il11 -1.21 0.343463

continued 111

Table 3.4 continued

E02 Interleukin 13 Il13 -1.38 0.230709

E03 Interleukin 13 receptor, alpha 1 Il13ra1 1.33 0.014745

E04 Interleukin 15 Il15 3.13 0.000410

E05 Interleukin 16 Il16 -1.20 0.436312

E06 Interleukin 17B Il17b -1.20 0.342296

E07 Interleukin 18 Il18 1.16 0.198855

E08 Interleukin 1 alpha Il1a 1.56 0.238539

E09 Interleukin 1 beta Il1b 1.47 0.671496

E10 Interleukin 1 family, member 6 Il1f6 1.01 0.918628

E11 Interleukin 1 family, member 8 Il1f8 1.04 0.575347

E12 Interleukin 1 receptor, type I Il1r1 1.16 0.240541

F01 Interleukin 1 receptor, type II Il1r2 1.66 0.561152

F02 Interleukin 20 Il20 -1.12 0.489555

F03 Interleukin 2 receptor, beta chain Il2rb 13.17 0.004233

Interleukin 2 receptor, gamma

F04 chain Il2rg 7.11 0.001782

F05 Interleukin 3 Il3 1.80 0.216321

F06 Interleukin 4 Il4 -1.47 0.257806

F07 Interleukin 5 receptor, alpha Il5ra -1.06 0.688428

F08 Interleukin 6 receptor, alpha Il6ra 1.19 0.343494

F09 Interleukin 6 signal transducer Il6st -1.19 0.338722

F10 Interleukin 8 receptor, beta Il8rb -1.04 0.554344

112 continued

Table 3.4 continued

F11 Integrin alpha M Itgam 1.96 0.004353

F12 Integrin beta 2 Itgb2 2.50 0.007309

G01 Lymphotoxin A Lta 2.29 0.241065

G02 Lymphotoxin B Ltb 1.74 0.026352

Macrophage migration inhibitory

G03 factor Mif -1.09 0.034021

Small inducible cytokine

G04 subfamily E, member 1 Scye1 -1.12 0.044669

G05 Secreted phosphoprotein 1 Spp1 2.06 0.034776

Transforming growth factor, beta

G06 1 Tgfb1 1.12 0.449473

G07 Tumor necrosis factor Tnf 16.68 0.001249

Tumor necrosis factor receptor

G08 superfamily, member 1a Tnfrsf1a 1.78 0.003348

Tumor necrosis factor receptor

G09 superfamily, member 1b Tnfrsf1b 2.59 0.019221

G10 CD40 ligand Cd40lg -1.51 0.227103

G11 Toll interacting protein Tollip -1.19 0.041760

G12 Chemokine (C motif) receptor 1 Xcr1 4.36 0.022416

H01 Glucuronidase, beta Gusb 1.39 0.011490

Hypoxanthine guanine Hprt1

H02 phosphoribosyl transferase 1 -1.20 0.004201

113 continued

Table 3.4 continued

Heat shock protein 90 alpha Hsp90ab

H03 (cytosolic), class B member 1 1 -1.13 0.028294

Glyceraldehyde-3-phosphate

H04 dehydrogenase Gapdh -1.14 0.000984

H05 Actin, beta Actb 1.12 0.193177

114

Table 3.5: Changes in inflammatory cytokine profile between rQnestin4.5 and

Asialo+rQnestin34.5 treated mice

Fold Change Gene Position Description (Asialo+rQnestin34.5 p-value Symbol /rQnestin34.5)

ATP-binding cassette, sub-family F A01 Abcf1 1.04 1.044636 (GCN20), member 1

A02 B-cell leukemia/lymphoma 6 Bcl6 -1.02 -1.019244

Chemokine (C-X-C motif) receptor A03 Cxcr5 -1.91 -1.914543 5

A04 Complement component 3 C3 -1.73 -1.725782

A05 Caspase 1 Casp1 -1.31 -1.313349

A06 Chemokine (C-C motif) ligand 1 Ccl1 -1.23 -1.229226

A07 Chemokine (C-C motif) ligand 11 Ccl11 -2.71 -2.707573

A08 Chemokine (C-C motif) ligand 12 Ccl12 -3.80 -3.796711

A09 Chemokine (C-C motif) ligand 17 Ccl17 1.31 1.307218

A10 Chemokine (C-C motif) ligand 19 Ccl19 -1.06 -1.064001

A11 Chemokine (C-C motif) ligand 2 Ccl2 -8.24 -8.239176

A12 Chemokine (C-C motif) ligand 20 Ccl20 -1.46 -1.463325

B01 Chemokine (C-C motif) ligand 22 Ccl22 1.06 1.059585

B02 Chemokine (C-C motif) ligand 24 Ccl24 -1.03 -1.031147

B03 Chemokine (C-C motif) ligand 25 Ccl25 1.11 1.111687

B04 Chemokine (C-C motif) ligand 3 Ccl3 1.09 1.085982

continued 115

Table 3.5 continued

B05 Chemokine (C-C motif) ligand 4 Ccl4 -2.84 -2.842184

B06 Chemokine (C-C motif) ligand 5 Ccl5 -2.80 -2.802569

B07 Chemokine (C-C motif) ligand 6 Ccl6 1.23 1.232212

B08 Chemokine (C-C motif) ligand 7 Ccl7 -7.66 -7.663482

B09 Chemokine (C-C motif) ligand 8 Ccl8 -2.49 -2.492338

B10 Chemokine (C-C motif) ligand 9 Ccl9 1.20 1.195613

B11 Chemokine (C-C motif) receptor 1 Ccr1 -1.00 -1.000520

B12 Chemokine (C-C motif) receptor 2 Ccr2 -1.33 -1.327075

C01 Chemokine (C-C motif) receptor 3 Ccr3 -1.17 -1.168777

C02 Chemokine (C-C motif) receptor 4 Ccr4 1.19 1.185914

C03 Chemokine (C-C motif) receptor 5 Ccr5 -1.02 -1.015718

C04 Chemokine (C-C motif) receptor 6 Ccr6 -1.16 -1.156488

C05 Chemokine (C-C motif) receptor 7 Ccr7 -1.47 -1.465609

C06 Chemokine (C-C motif) receptor 8 Ccr8 1.02 1.021012

C07 Chemokine (C-C motif) receptor 9 Ccr9 -1.59 -1.591073

C-reactive protein, pentraxin- C08 Crp -1.09 -1.094673 related

Chemokine (C-X3-C motif) ligand C09 Cxc3cl1 1.00 1.002255 1

C10 Chemokine (C-X-C motif) ligand 1 Cxcl1 -1.34 -1.343503

Chemokine (C-X-C motif) ligand - C11 Cxcl10 -15.85 10 15.853726

continued 116

Table 3.5 continued

Chemokine (C-X-C motif) ligand C12 Cxcl11 -9.74 -9.743905 11

Chemokine (C-X-C motif) ligand D01 Cxcl12 1.02 1.017303 12

Chemokine (C-X-C motif) ligand D02 Cxcl13 -3.15 -3.149778 13

Chemokine (C-X-C motif) ligand D03 Cxcl15 1.03 1.025445 15

D04 Platelet factor 4 Pf4 1.69 1.687339

D05 Chemokine (C-X-C motif) ligand 5 Cxcl5 -1.81 -1.809072

D06 Chemokine (C-X-C motif) ligand 9 Cxcl9 -7.48 -7.479802

Chemokine (C-X-C motif) receptor D07 Cxcr3 -1.08 -1.076054 3

D08 Chemokine (C-C motif) receptor 10 Ccr10 1.08 1.077733

- D09 Interferon gamma Ifng -10.35 10.351371

D10 Interleukin 10 Il10 -1.28 -1.280760

D11 Interleukin 10 receptor, alpha Il10ra -1.36 -1.361314

D12 Interleukin 10 receptor, beta Il10rb 1.03 1.033652

E01 Interleukin 11 Il11 1.02 1.018715

E02 Interleukin 13 Il13 -1.73 -1.728475

E03 Interleukin 13 receptor, alpha 1 Il13ra1 -1.03 -1.027936

continued 117

Table 3.5 continued

E04 Interleukin 15 Il15 -1.86 -1.858643

E05 Interleukin 16 Il16 1.22 1.224549

E06 Interleukin 17B Il17b -1.27 -1.273236

E07 Interleukin 18 Il18 -1.09 -1.088242

E08 Interleukin 1 alpha Il1a 1.40 1.403715

E09 Interleukin 1 beta Il1b -1.00 -1.003124

E10 Interleukin 1 family, member 6 Il1f6 1.03 1.025445

E11 Interleukin 1 family, member 8 Il1f8 1.03 1.025445

E12 Interleukin 1 receptor, type I Il1r1 1.12 1.122721

F01 Interleukin 1 receptor, type II Il1r2 -1.33 -1.325007

F02 Interleukin 20 Il20 1.03 1.025445

F03 Interleukin 2 receptor, beta chain Il2rb -3.34 -3.339773

F04 Interleukin 2 receptor, gamma chain Il2rg -2.52 -2.516642

F05 Interleukin 3 Il3 -1.08 -1.076800

F06 Interleukin 4 Il4 -2.32 -2.320603

F07 Interleukin 5 receptor, alpha Il5ra 1.12 1.116320

F08 Interleukin 6 receptor, alpha Il6ra 1.09 1.085606

F09 Interleukin 6 signal transducer Il6st 1.14 1.142940

F10 Interleukin 8 receptor, beta Il8rb -1.49 -1.494331

F11 Integrin alpha M Itgam -1.11 -1.105348

F12 Integrin beta 2 Itgb2 -1.31 -1.311529

G01 Lymphotoxin A Lta 1.04 1.040841

continued 118

Table 3.5 continued

G02 Lymphotoxin B Ltb -1.32 -1.322255

Macrophage migration inhibitory G03 Mif 1.06 1.059218 factor

Small inducible cytokine subfamily G04 Scye1 1.15 1.152686 E, member 1

G05 Secreted phosphoprotein 1 Spp1 1.08 1.082037

G06 Transforming growth factor, beta 1 Tgfb1 -1.07 -1.066031

G07 Tumor necrosis factor Tnf -3.07 -3.066834

Tumor necrosis factor receptor G08 Tnfrsf1a -1.24 -1.236918 superfamily, member 1a

Tumor necrosis factor receptor G09 Tnfrsf1b -1.45 -1.449947 superfamily, member 1b

G10 CD40 ligand Cd40lg 2.42 2.417473

G11 Toll interacting protein Tollip 1.08 1.078854

G12 Chemokine (C motif) receptor 1 Xcr1 -1.92 -1.922522

H01 Glucuronidase, beta Gusb 1.13 1.128964

Hypoxanthine guanine H02 Hprt1 1.03 1.033114 phosphoribosyl transferase 1

Heat shock protein 90 alpha Hsp90a H03 1.02 1.017656 (cytosolic), class B member 1 b1

Glyceraldehyde-3-phosphate Gapdh H04 1.07 1.071402 dehydrogenase

continued 119

Table 3.5 continued

H05 Actin, beta Actb -1.27 -1.271693

120

Chapter 4: NK cells kill oHSV infected glioblastoma cells using NKp30 and NKp46

Introduction

The findings in chapter 3 that NK cell activation and recruitment to the site of oHSV are deleterious to OV efficacy build upon extensive previous findings that innate antiviral immunity is a barrier to successful HSV virotherapy for glioblastoma9,86,90,118,181,195.

Based on these findings, it was necessary to decipher the key signal that lead not only to

NK mediated clearance of glioblastoma cells but also how this processes changes following infection of these tumors with oncolytic HSV. Several studies have evaluated and confirmed the ability of NK cells to kill tumors derived from the CNS107,400.

Additionally, NK cells have been documented to mediate the clearance of HSV-infected target cells112. However, to date there has not been thorough examination regarding the key mechanistic signals that mediate NK cell recognition and clearance of oHSV infected glioblastoma.

The results from this work are particularly useful within the field of oncolytic viral therapy. Despite the conflicting view points that NK cells are a benefit or a hindrance to

OV efficacy, most investigators would argue that NK cells and their place within innate immunity have a critical role in achieving success with this therapeutic modality369. As a result, the knowledge gained from uncovering the mechanistic signals governing NK

121 mediated recognition of OV infected cells has the potential of benefiting both schools of thought. For instance, in instances where the NK response is deleterious to OV efficacy, it may be necessary to design oncolytic viruses that express decoys or suppressors of NK activating ligands or combine the oncolytic virus with co-therapies that achieve similar

NK avoidance of oHSV infected target cells. In tumors where NK mediated killing is beneficial to OV therapy, the opposite approach could be used so that the oncolytic virus is combined with an immunostimulatory agent that heightens the expression of critical

NK activating ligands. In either case, it will be necessary to uncover the underlying signals that mediate NK recognition of OV infected tumor cells.

NK cells possess a variety of receptors, such as NKG2D, DNAM-1, and the natural cytotoxicity receptors (NCR) NKp30, NKp44, and NKp46, that mediate NK cytotoxic functions; however, the key receptor—ligand interactions that coordinate this response is largely unresolved. NKG2D and DNAM-1 are well-characterized activating receptors and previous studies evaluating the role of NK mediated killing of glioblastoma stem cells implicated DNAM-1, but not NKG2D, as a critical receptor in this process107. It is important to note, however, that the relative importance of each receptor is specific to the type of tumor being considered. In the context of NK recognition of cells infected with wild type HSV, the natural cytotoxicity receptors were essential for viral clearance112.

With these studies as a foundation, we have been able to confirm that oHSV infection of glioblastoma cells induces a series of cell surface and intracellular signaling changes in

122 both infected glioblastoma and responding NK cells that are responsible for mediating

NK preferential clearance of virally infected glioblastoma. These findings provide the first mechanistic description of oHSV-induced signaling within NK cells and infected glioblastomas that leads to the preferential lysis of oHSV infected tumor cells. With the understanding of these initial mechanistic responses to OV therapy, future novel therapeutic approaches can be pursued that target these specific components of antiviral

NK signaling.

Methods

Cell culture

The glioblastoma cells used include human glioblastoma cell lines (U87dEGFR, U251, and Gli36dEGFR) and primary human glioblastoma cells enriched for stem-like cell properties (X12401). These cells were cultured in Dulbeco’s Modified Eagle’s Medium

(Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, penicillin (100

U/ml), and streptomycin (100 µg/ml). K562 CML derived cells, human and mouse derived NK cells were cultured in RPMI-1640 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, penicillin (100 U/ml), and streptomycin (100 µg/ml). Cells

o were cultured at 37 C supplemented with 5% CO2.

NK cell isolation

NK cells or cytotoxic T lymphocytes (CTLs) were enriched from peripheral blood leukopacks of healthy donors (American Red Cross, Columbus, Ohio) using RosetteSep

123 cocktail (StemCell Technologies, Vancouver, Canada). The enriched NK cells were then further purified using positive selection CD56 magnetic bead sorting while the enriched

CTLs were purified using positive selection CD8 magnetic bead sorting (Miltenyi

Biotec). Murine NK cells were isolated from mouse splenocytes. Splenocytes were isolated and cultured with DX5-PE anti-mouse antibody (BD). After washing the cells, anti-PE MicroBeads (Miltenyi Biotec) were added and the resulting murine NK cells were isolated using positive selection bead sorting. Once NK cells or CTLs were isolated, they were cultured overnight in 10% RPMI prior to being used in downstream applications.

Cytotoxicity assay

The panel of human glioblastoma cells was plated overnight at 104 cells/well. The following day, cells were infected for 8 hours with rQnestin34.5 (MOI 1.0) or mock infected. Human NK cells or CTLs isolated as described above were added at different effector:target ratios in the presence of human IL-15 (for NK cells) or IL-2 (for CTLs)

(Miltenyi Biotec). The co-culture was allowed to proceed for 4 hours at 37oC.

Glioblastoma lysis was assessed by measuring glucose 6-phosphate dehydrogenase released from lysed cells using the Vybrant Cytotoxicity Assay Kit (Molecular Probes,

Eugene, OR)402. Supernatants were collected for the detection of IFN-γ by enzyme- linked immunosorbent assay110. Granzyme B production from co-cultured NK cells was evaluated using ELISPOT403. For studies evaluating pharmacological inhibitors of cytolysis, we preincubated NK cells for 1 hour with: EGTA/Mg2+ (2mM/4mM),

124

Chloroquine (100 µg/mL), cyclosporine A (5 ug/mL), emetine (10 ug/mL), or Fas-Fc (10 ug/mL), before adding to our infected/mock infected panel of human glioblastoma. The co-culture was allowed to proceed for 4 hours at 37oC. Glioblastoma lysis was then assessed as described above. For NCR masking experiments on human our mouse NK cells, NK cells were preincubated for 1 hour with the appropriate blocking antibody.

Human NK cells were treated with the following IgM blocking antibodies: DNAM-1

(F5), NKp30 (F252), NKp44 (KS38), and NKp46 (KL247) or the IgG antibody against

NKG2D (BD)404 or MHC-I (W6/32). Murine NK cells were treated with blocking antibody against murine NKp46405. NK cells were then added to our infected/mock infected panel of human glioblastoma. The co-culture was allowed to proceed for 4 hours at 37oC.

Analysis of NK ligand expression

Recombinant human NKp30-IgG or NKp46-IgG fusion proteins (R & D Systems,

Minneapolis, MN) was used to investigate the expression of NK ligands on glioblastoma

8 hours following oHSV infection, glioblastoma treated with temozolomide (200 µM), 10 gy radiation, or hypoxia (1% O2, 5% CO2). Staining of NK ligands was performed by the addition of 10 µl of reconstituted IgG-fusion proteins to 105 glioblastoma cells in a

100 µl volume and incubated for 2 hours on ice. After washing with PBS, cells were incubated with APC-conjugated mouse anti-human IgGfc (Jackson ImmunoResearch,

West Grove, PA) at a 1:100 dilution and incubated for 20 minutes. We also analyzed the expression of B7-H6 following oHSV infection of glioblastoma. Eight hours after

125 infection, cells were cultured with 10 µg/mL of mouse anti-human B7-H6 (gift from

ZymoGenetics, Seattle, WA) for 30 minutes. After washing with PBS, cells were incubated with APC-anti-mouse IgG (Invitrogen) for 20 minutes. Cell surface staining of oHSV infected or mock infected glioblastoma was also assessed with the following anti- human antibodies: MICA/B-PE, ULBP-1-PE, ULBP-2-PE, CD155-PE (R&D); HLA-

ABC-PE (eBioscience); and CD48-PE, CD112-PE (Biolegend). Lastly, following antibody staining, cells were re-suspended in 1% formalin and analyzed using a FACS

Calibur (Becton Dickinson, Mountain View, CA).

Results

Human glioblastoma activate NK cells from human donors

We first compared the ability of U87dEGFR human glioblastoma cells and K562 CML cells, a prototypical NK target cell, to activate human derived NK cells. Using an

ELISPOT assay to detect granzyme B or IFN-γ production, NK cells were equivalently activated following an overnight co-culture with either cell type (Figure 4.1a-b). Since we confirmed that human glioblastoma was able to endogenously mediate NK activation, we tested how the infection of these tumors with oHSV would modulate this NK activation state. Co-culture of oHSV-infected U87dEGFR cells with NK cells produces significantly more IFN-γ than NK cells co-cultured with uninfected GBM (Figure 4.1c) and more granzyme B expression (Figure 4.1d).

126

NK cells preferentially kill oHSV infected glioblastoma in both mouse and human models

Because we had observed increased expression of IL-15 transcript in tumors following oHSV administration (Figure 4.2), we used IL-15 activated human NK cells in our in vitro experiments. Studies from our laboratory have also demonstrated that IL-15 is a critical monokine responsible for promoting NK cell survival, proliferation, development, and activation406-411. IL-15 “primes” NK cells for IFN-γ production when exposed to IL-

1b, IL-12 or IL-18. One mechanism that has been shown is through the induction of T-bet expression while simultaneously shutting down inhibitory TGF-β signaling110. IL-15 also activates NK cells to kill virally infected cells and tumor cells.

IL-15 activated NK cells were able to readily kill uninfected human glioblastoma and killing was enhanced when the panel of glioblastoma was first infected with oHSV

(Figure 4.3a). Unstimulated NK cells also achieved the same effect, albeit to a much reduced level (Figure 4.3b). Confirming our human in vitro co-culture data, murine NK cells also preferentially cleared oHSV infected glioblastoma (Figure 4.3c). Lastly, we evaluated NK mediated killing at various time points after oHSV infection and found that

NK cells preferentially lysed virally infected over mock-infected GBM cells at all tested time points (Figure 4.4). Notably, this antiviral response was unique to NK cells as CTLs from human donors were unable to recapitulate this cytotoxicity profile (Figure 4.5). As a control, we confirmed that 12 hours post-infection at various multiplicities of infection, viral infection was not leading to cell death in any cell line tested.

127

Figure 4.6 also suggests that increases in NK mediated cytotoxicity are occurring at least in part through the capacity of IL-15 to drastically induce the expression of NCR, especially NKp30. Notably, exogenous TFG-β was able to abrogate the increased expression of NKp30 (Figure 4.6). This is relevant since NKp30 functions as a critical activating receptor on NK cells. Taken together, these findings demonstrate that GBM infection with oHSV leads to preferential activation of NK cells with a concomitant increase in NK-mediated lysis of oHSV infected cells compared to uninfected GBMs.

NK mediated killing of oHSV infected glioblastoma is dependent upon cell contact, perforin, and DNAM-1

In order to explore the mechanism for NK mediated clearance of oHSV infected glioblastoma (including a freshly excised human glioblastoma grown to enhance the brain tumor stem cell population, X12), we confirmed that: a) maximum killing was cell contact dependent, b) killing was likely dependent on perforin412-414, and c) NK mediated lysis was independent of MHC-I blockade (Figure 4.7). For our pharmacological inhibitor studies, we used a panel of cytolytic signaling inhibitors (target of inhibitor):

4mM EGTA (perforin)412, 100 ug/mL chloroquine (perforin)413, 5 ug/mL cyclosporine A

(TNF-family)415-417, 10 ug/mL emetine (FasL and NKp30)415-417, 10 ug/mL Fas-Fc

(FasL/Fas)415-417. However, maximum inhibition was achieved with EGTA and chloroquine, indicating that perforin was the critical mechanism leading to NK mediated killing (Figure 4.8).

128

We further evaluated HLA-ABC expression on our glioblastoma panel and found that surface staining was either unchanged or modestly downregulated following oHSV infection (Figure 4.9a). Ligands for the canonical NK activating receptors DNAM-1 and

NKG2D were endogenously expressed on our glioblastoma panel yet were unchanged following viral infection (Figure 4.9b). Blockade of the NKG2D receptor on our glioblastoma panel inhibited killing in ¼ glioblastoma cells, but analyzed collectively this was not significant (P=0.34) (Figure 4.7); however, blockage of DNAM-1 achieved moderate inhibition of glioblastoma clearance in both oHSV (P=0.03) and mock infected cells (P=0.01) (Figure 4.9c). Collectively, these findings thus indicated that human (and mouse) NK cells are able to kill oHSV-infected (and to a lesser degree, uninfected) glioblastoma cells utilizing perforin-based mechanisms and canonical NK cell receptor recognition of oHSV-infected tumor cells.

The NCRs NKp30 and NKp46 mediate clearance of oHSV-infected glioblastoma.

Since the mechanism by which NK cells and virus interact with each other is largely unknown, we tried to determine if non-canonical NCR mediated killing was the critical mechanism accounting for clearance of oHSV infected glioblastoma107,115. NK mediated clearance was significantly inhibited by blocking either NKp30 (P=0.003) or NKp46

(P=0.02), while NKp44 blockade alone was not significant (P=0.41) (Figure 4.10). We also extended the relevancy of these findings to a murine in vitro model where murine

NKp46 (the sole NCR present in mice) was found to mediate NK mediated killing of oHSV infected glioblastoma (Figure 4.11).

129

Since the cellular ligands for NCRs are largely unknown and their identify represent an area of intense investigation, we used commercially available NCR fusion proteins

NKp30-Ig and NKp46-Ig to detect ligand expression on our glioblastoma panel115. With this approach, we detected enhanced NKp30-Ig and NKp46-Ig staining following oHSV infection (Figure 4.12a). Additionally, since rQnestin34.5 expresses GFP as a marker of viral infection, this allowed us to test for NCR ligand staining specifically within virally infected (GFP+) cells. For both NKp30-Ig and NKp46-Ig, we detected maximal ligand staining within the virally infected (GFPhigh) cell population (Figure 4.12b). Notably, neither NKG2D-Ig nor NKp44-Ig fusion proteins detected enhanced ligand staining following oHSV infection. Additionally, detection of ligands with either of these fusion proteins did not exhibit the characteristic staining pattern within the GFPhigh population that was seen with NKp30-Ig and NKp46-Ig (Fig 4.12b).

A recent report demonstrated that B7-H6 is a novel ligand for NKp30418. Consequently, we tested whether this protein was involved in this context. Although B7-H6 expression is modestly expressed in the panel of glioblastoma cells, viral infection was unable to modulate its cell surface staining (Figure 4.12c).

Lastly, we evaluated if NKp30 and NKp46 ligand expression were up-regulated solely after oHSV infection or if various modes of cellular stress could induce them. Exposing glioblastoma to either radiation or hypoxia induced moderate expression of NKp30

130 ligand but not NKp46 ligand (Figure 4.13). However, temozolomide (TMZ) treatment accounted for the most robust induction of NKp30 and NKp46 ligand expression to a magnitude similar to that observed with oHSV infection (Figure 4.13). Therefore, these findings demonstrate that NK mediated clearance of oHSV-treated glioblastoma is coordinated primarily through the NK receptors NKp30 and NKp46 that recognize cognate ligands on tumors up-regulated by oHSV or TMZ.

Discussion

For the first time, we have discovered that oHSV-infected human glioblastoma up- regulate ligands for the NK cell natural cytotoxicity receptors, NKp30 and NKp46.

Recognition of these ligands is a critical mechanism for NK cell-mediated clearance of oHSV-infected cells. This study thus demonstrates that the observed rapid host NK cell response to glioblastoma following oHSV therapy functions to cooperate with other innate immune effector populations to reduce, rather than enhance, the efficacy of oHSV treatment via the NCRs, NKp30/NKp46. Finally, we argue that clinical reports of patients with rare deficiencies in NK cell function being susceptible to HSV1 encephalitis104-106 lead us to conclude the validity of the in vitro and in vivo model systems employed to answer the overarching hypothesis that NK cell activation is deleterious to OV efficacy.

A common feature of glioblastoma is the endogenous expression of varying amounts of ligands for the activating NK cell receptors107,115. Based on this ligand expression profile,

131

NK cells have been demonstrated to actively lyse glioblastoma cells through the NKp46 and DNAM-1 receptors107. Recent studies have aimed to elucidate the role of HSV-1 infection in modulating NK activating ligand expression. Consistent with our results,

HSV-1 infection resulted in increased susceptibility to NK mediated lysis in a NKG2D independent, NCR dependent manner112. The family of NCRs consist of NKp46 which is endogenously expressed on NK cells and NKT-like cells in both mice and humans405;

NKp44 which is solely expressed on human NK cells, with constitutive expression only after cytokine stimulation, and plasmacytoid dendritic cells419; and NKp30 which is exclusively expressed on resting and activated human NK cells420.

The novel and significant finding is that human glioblastomas, infected with oHSV, up- regulate endogenous ligands recognized most significantly and consistently, by human

NKp46/NKp30 expressed on human NK cells. Mice models, where NKp46 is present, confirm the relevance of the NK cell response against infected murine and human glioblastomas. The observed mechanism in human cells is specific to NKp46 and

NKp30: figure 4.10 shows blockade of these two receptors significantly and consistently reduces NK cell cytotoxicity against glioblastoma cells more so than blockade of NKp44,

NKG2D, or DNAM1 (Figures 4.7, 4.9c, 4.10). In addition, the histograms of figure

4.12a show that the intensity of NKp30 and NKp46 ligands, respectively, is increased with viral infection, thus indicating that these ligands are better recognized. Finally, figure 4.13 shows that the observed increase in ligand expression in tumor cells is absent when cells are exposed to non-specific methods of tumor killing (hypoxia or radiation);

132 rather, expression of these NK ligands occurs more specifically with either OV infection or temozolomide. The sum of this data argues for relative specific increased recognition of the ligands for NKp30 and NKp46.

While the identity of ligands for NCRs is a field of intense investigation421, recent discoveries have identified influenza hemagglutinin (HA) as an activating ligand for

NKp46 and NKp44422. Although HA is a target for neutralizing antibodies and is therefore subject to evolutionary pressures to evade immunity, NKp46 and NKp44 are consistently able to recognize various HA subtypes despite antigenic drift421. Besides mediating the eradication of tumor and virally infected cells, NKp30 interacts with immature dendritic cells (imDCs). Following NKp30 binding to an unknown ligand on imDCs423, the imDCs are subsequently either killed or develop into mature DCs that can mediate a Th1 response424 culminating in tumor/viral eradication. To counteract this process, human cytomegalovirus tegument protein pp65 impedes NKp30 activation through NKp30-CD3ζ receptor dissociation and the concomitant circumvention of

NKp30 mediated maturation of dendritic cells425,426. Known activating ligands for

NKp30 include two activating cellular proteins. These include B7-H6418, a cell surface protein associated with tumor formation and Bat3427, a released cellular stress protein.

While NKp30 and NKp46 were critical mediators of NK mediated lysis of oHSV infected glioblastoma, B7-H6 did not appear to be involved in this response. Moreover, cell surface expression of NKp30/NKp46 ligands were expressed at similar levels following either oHSV infection or treatment with TMZ.

133

These findings raise several interesting points including whether the ligands for NKp30 and NKp46 are of viral or cellular origin. We detected these ligands predominantly in our GFPhigh population of oHSV infected glioblastoma, suggesting that the unknown ligand is primarily expressed in virally infected cells. However, various stressors, in particular TMZ, were able to similarly induce NKp30/NKp46 ligand expression. Taken together, our findings suggest that oHSV infection and the cellular stress imparted by

TMZ result in similar signals culminating in the expression of these NK ligands. Once the identity of the NKp30 and NKp46 ligand is uncovered, future work will investigate whether they can be utilized to predict sensitivity to oncolysis for personalized treatment.

Additionally, the finding that TMZ induces NKp30/NKp46 ligand expression raises interesting clinical implications. For instance, it may be useful to combine standard TMZ therapy with an immunostimulatory agent that can elicit a robust NK anti-tumor response within the tumor microenvironment. Combining TMZ with oHSV has been shown to have synergistic effects428. The findings in this chapter suggest that if initial anti-viral immunity can be evaded, downstream anti-tumor efficacy can potentially be achieved with this co-therapy. Lastly, since the expression of NCRs is coordinated as either

NCRbright or NCRdim429, future studies will need to examine if NK cells exhibit an

NCRbright phenotype following oHSV treatment in humans and if this expression pattern correlates with diminished oncolytic viral replication.

134

The results from this study outline several important advances within the field of oHSV therapy for glioblastoma. Lastly, by clarifying the critical receptor—ligand interactions that mediate human NK clearance of glioblastoma infected with oncolytic HSV, we have uncovered a therapeutic target that, when blocked, could limit NK clearance of these infected cells. By selectively targeting this cellular response in future clinical trials, we identified a novel avenue for potentially achieving successful oHSV therapy.

135

A. continued

Figure 4.1: NK cells are activated following oncolytic viral infection of co-cultured target cells.

(A) A total of 104 cells/well of U87dEGFR glioblastoma cells or K562 multiple myeloma cells were plated. Enriched human NK cells were subsequently co-cultured overnight at varying effector:target rations in the presence of IL-2. The following day granzyme B production was assessed with an ELISPOT assay. (B) Cells were plated and co-cultured as described in (A). IFN-γ production was then assessed with an ELISPOT assay. (B)

U87dEGFR cells were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) or mock-infected for 8 hours. Enriched human NK cells were subsequently co-cultured at an effector:target ratio of 12:1 in the presence of IL-15. Four hours later granzyme B production was assessed with an ELISPOT assay. (D) U87dEGFR cells were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) or mock-infected for 8 hours.

Enriched human NK cells were subsequently co-cultured at an effector:target ratio of

12:1 in the presence of IL-15. Four hours later, supernatant was collected and analyzed using an IFN-γ ELISA assay. Error bars represent +/- standard deviation. 136

Figure 4.1 continued

B.

C. continued

137

Figure 4.1 continued

D.

138

Figure 4.2: IL-15 is expressed following oncolytic viral infection.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice

(n=4-5/group) and allowed to grow for 7 days. Mice were then depleted of their NK cells with a single intraperitoneal injection of anti-Asialo-GM1 mAb (50µl antibody+50µl

4 H2O). Two days later, rQnestin34.5 (10 pfu/3µl) or vehicle was inoculated at the site of tumor implantation. Mice were sacrificed 72 hours later for mRNA isolation of the tumor bearing hemisphere. Following mRNA conversion to cDNA, the expression of IL-15 was evaluated with RT-qPCR. *** P < 0.001. Error bars represent +/- standard deviation.

139

A. continued

Figure 4.3: NK cells preferentially kill glioblastoma cells infected with oncolytic virus.

(A) A panel of human glioblastoma cell lines was plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) or mock-infected for 8 hours. Enriched human NK cells were co-cultured at varying effector:target ratios in the presence of IL-15. Four hours later glioblastoma lysis was assessed. (B) Cells were plated, treated, and assessed for lysis as described in (A) but in the absence of IL-15. (C) The ability of NK cells from athymic mice or C57BL/6 wild type mice to lyse U87dEGFR or KR158dEGFR glioblastomas, respectively, was assessed. NK cells isolated from mouse splenocytes were enriched and co-cultured at varying effector:target ratios in the presence of IL-15.

Four hours later glioblastoma lysis was assessed. Error bars represent +/- standard deviation. E = effector: T= Target

140

Figure 4.3 continued

B.

C.

141

Figure 4.4: NK cells preferentially kill virally infected glioblastoma cells at early timepoints after infection.

Gli36dEGFR glioblastoma cells were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) for various amount of time prior to adding oncolytic virus. In either the presence or absence of IL-15, enriched human NK cells were co-cultured with either infected or uninfected target cells at an effector:target ratio of 12:1. Four hours later glioblastoma lysis was assessed. Error bars represent +/- standard deviation.

142

Figure 4.5: Cytotoxic T lymphocytes and NK cells exhibit a differential cytotoxic profile against virally infected glioblastoma cells.

U251 and Gli36dEGFR human glioblastoma cell lines were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) or mock-infected for 8 hours. Enriched human cytotoxic T lymphocytes were subsequently co-cultured at an effector:target ratio of 15:1 in the presence of IL-2. Four hours later glioblastoma lysis was assessed. ** P < 0.01.

Error bars represent +/- standard deviation.

143

Figure 4.6: Exogenous IL-15 and TGF-β inversely regulate NKp30 expression.

Human NK cells were enriched with CD56 bead sorting and cultured overnight in the presence of either IL-15 or IL-15 and TGF-β. The following day, NKp30 expression was assessed by FACS.

144

Figure 4.7: NK killing oHSV infected glioblastoma requires cell contact but occurs independently of NKG2D or MHC I modulation.

A panel of human glioblastoma cell lines was plated and infected with rQnestin34.5

(MOI 1.0) for 8 hours. Prior to adding enriched human NK cells to the glioblastoma co- culture at an effector:target ratio of 12.5:1, they were pre-treated as follows: mock treated and placed in direct contact with glioblastoma; separated by a 0.4 µm transwell insert; treated with either 2mM/4mM EGTA/Mg2+ to chelate calcium or 100 µg/mL chloroquine to prevent granule acidification to inhibit perforin mediated killing; or treated with 10

µg/mL of either anti-MHC-I or anti-NKG2D blocking antibody. Four hours later in the presence of IL-15, glioblastoma lysis was assessed. * P < 0.05; ** P < 0.01; *** P <

0.001. Error bars represent +/- standard deviation. 145

Figure 4.8: NK cell killing of virally infected cells occurs primarily through a perforin mediated mechanism.

A panel of human glioblastoma cell lines (U87dEGFR, Gli36dEGFR, or U251) or primary human glioblastoma cells enriched for stem-like cell properties (X12) were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) or mock-infected for 8 hours. Enriched human NK cells were subsequently pretreated for 1 hour with one of the following cytolytic signaling inhibitors: 100 ug/mL chloroquine, 4mM EGTA, 10 ug/mL emetine, 10 ug/mL Fas-Fc, 5 ug/mL cyclosporine A. Following pretreatment, NK cells were subsequently co-cultured with either infected or uninfected target cells at an effector:target ratio of 12:1 in the presence IL-15. Four hours later glioblastoma lysis was assessed.

146

A. continued

Figure 4.9: Ligands for NK cells are endogenously expressed on glioblastoma.

(A) A panel of human glioblastoma cell lines (U87dEGFR, Gli36dEGFR, or U251) or primary human glioblastoma cells enriched for stem-like cell properties (X12) were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) for 8 hours. Cells were collected with EDTA and incubated with an HLA-ABC specific antibody so that surface expression could be assessed with FACS. (B) U87dEGFR and Gli36dEGFR cells were plated and infected with rQnestin34.5 (MOI 1.0) for 8 hours. Cells were collected with

EDTA and incubated with a panel of antibodies for the known ligands of NK activating receptors so that surface expression could be assessed with FACS. (C) The panel of glioblastoma cells was plated and infected as described in (A). Enriched human NK cells were pretreated with anti-DNAM-1 blocking antibody prior to co-culture with infected glioblastoma. Four hours later in the presence of IL-15, glioblastoma lysis was assessed.

* P < 0.05.

147

Figure 4.9 continued

B.

C.

148

Figure 4.10: Human NK cells clear oHSV infected glioblastoma through NKp30 and

NKp46.

A panel of human glioblastoma cell lines (U87dEGFR, Gli36dEGFR, or U251) or primary human glioblastoma cells enriched for stem-like cell properties (X12) were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) for 8 hours. Enriched human NK cells were treated with blocking antibodies against NKp30, NKp44, or

NKp46 prior to adding to the glioblastoma co-culture at an effector:target ratio of 12.5:1.

Four hours later in the presence of IL-15, glioblastoma lysis was assessed. * P < 0.05; **

P < 0.01; *** P < 0.001. Error bars represent +/- standard deviation.

149

Figure 4.11: Murine NK cells kill virally infected glioblastoma cells in an NKp46 dependent manner.

U87dEGFR and KR158dEGFR glioblastoma cells were plated (104 cells/well) and infected with rQnestin34.5 (MOI 1.0) for 8 hours. Enriched murine NK cells were treated with 10 µg/mL anti-NKp46 blocking antibody and then added to the glioblastoma co-culture at an effector:target ratio of 15:1. Four hours later in the presence of IL-15, glioblastoma lysis was assessed. * P < 0.05. Error bars represent +/- standard deviation.

150

A. continued

Figure 4.12: OV infection of glioblastoma induces novel ligands for NKp30 and NKp46.

(A) Human glioblastoma cells were infected with rQnestin34.5 (MOI 1.0) for 8 hours.

Cells were collected with EDTA and incubated with either NKp30-IgG or NKp46-IgG for 2 hours. Cells were subsequently washed and cultured with an APC-conjugated secondary antibody and then analyzed by FACS. (B) U87dEGFR cells were infected, incubated with either NKp30-IgG or NKp46-IgG, and cultured with an APC-conjugated secondary antibody as described in (A). Since rQnestin34.5 expressed GFP following infection, this allowed for the co-localization of NKp30 or NKp46 ligand expression in virally infected cells. (C) Cells were plated and infected as described in (A). Cells were collected with EDTA and cultured with anti-B7-H6 antibody for 2 hours. Cells were subsequently washed and cultured with an APC-conjugated secondary antibody and then analyzed by FACS.

151

Figure 4.12 continued

B.

C.

152

A. continued

Figure 4.13: Induction of NCR expression occurs after oHSV infection and TMZ exposure but not other forms of cellular stress

(A) U87dEGFR and X12 glioblastoma were exposed to various forms of cellular stress

(10 gy radiation, hypoxia, or 200 µM temozolomide). Cells were collected with EDTA and incubated with either NKp30-IgG or NKp46-IgG for 2 hours. Cells were subsequently washed and cultured with an APC-conjugated secondary antibody and then analyzed by FACS. (B) The cell surface expression of NKp30 and NKp46 ligand was compared following infection with rQnestin34.5 or exposure to radiation, hypoxia, or temozolomide.

153

Figure 4.13 continued

B.

154

Chapter 5: The innate immune response is attenuated by valproic acid co-administration

with oHSV

Introduction

Based on the work from chapter 3 and previous findings from our laboratory, the innate immune response has a deleterious impact on achieving successful OV therapy. In particular, the activated response of both NK cells and microglia/macrophages to oHSV infection impedes viral infection, replication, and ultimate tumor lysis86,90. Various approaches have been used to circumvent this barrier, including the use of tumor- trafficking immune cells430,431, second generation oncolytic viruses expressing immuno- evasive proteins389, and pharmacologic adjuvants that suppress the antiviral immune response90,116-119,144,327,390,432.

In order to circumvent antiviral immunity, we have demonstrated that pharmacological modulation of the innate immune response by administration of cyclophosphamide prior to oHSV treatment resulted in enhanced OV efficacy as measured by viral replication and overall animal survival118,390. We proceeded to evaluate IFN-γ production 72 hrs after rat glioblastomas were treated with oHSV and showed that protein levels of IFN-γ were robustly increased following oHSV administration; however, CPA pretreatment was able to attenuate this response90. Additionally, treatment with 4-hydroxyperoxy-CPA (4HC),

155 the intermediate active metabolite of CPA, impaired monokine-induced IFN-γ production by NK cells90. Taken together, these findings demonstrate that pharmacological suppression of innate immunity in the context of OV therapy, as seen with CPA, is a mechanism towards achieving therapeutic success.

In addition to CPA, valproic acid has been used by both our laboratory and others to enhance OV therapy. VPA is a commonly used antiepileptic agent with histone deacetylase inhibitory functions. VPA and other HDAC inhibitors have been found to have an anticancer activity308,309,433. Moreover, HDAC inhibitors have been shown to upregulate the transcriptional activity of virally delivered transgenes both in vitro and in vivo434-436 and to prevent the transcriptional activation of IFN-stimulated genes (ISGs) in response to IFN treatment and viral infection. These features make HDAC inhibitors an attractive co-therapy to augment HSV based oncolytic virotherapy. Our laboratory and others have previously demonstrated that pretreatment with VPA can upregulate the transcriptional activity of HSV genes, limit the antiviral effects of IFN by inhibiting the induction of ISGs, and enhance the in vivo therapeutic efficacy of oncolytic HSV 119.

However, to-date an examination into the ability of VPA to modulate the innate immune response to oHSV infection is lacking. By understanding the ability of VPA to alter the kinetics of immune cell recruitment, activation, and inflammation, we will uncover insights that can be harnessed for designing oHSV co-therapy regimens. In addition, the

156 ability of VPA to modulate immune cell activation states will have clinical applicability beyond the field of virotherapy based on its use as an antineoplastic agent.

Methods

Cell culture

The glioblastoma cells used include human glioblastoma cell lines (U87dEGFR, U251, and Gli36dEGFR) and primary human glioblastoma cells enriched for stem-like cell properties (X12401). These cells were cultured in Dulbeco’s Modified Eagle’s Medium

(Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, penicillin (100

U/ml), and streptomycin (100 µg/ml). Donor derived NK cells and the NK-92 cell line were cultured in RPMI-1640 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, penicillin (100 U/ml), and streptomycin (100 µg/ml). Cells were cultured

o at 37 C supplemented with 5% CO2.

Animal studies

Athymic mice (Charles River Laboratories, Wilmington, MA) were anesthetized by intraperitoneal administration of ketamine (100mg/kg)/xylazine (20mg/kg) and stereotactically injected with 1 x 105 human U87dEGFR glioblastoma cells into the right frontal lobe of the brain (2mm lateral and 1mm anterior to bregma at a depth of 3 mm).

For VPA (Sigma-Aldrich, St Louis, MO) administration, drug was administered intraperitoneally (300mg/kg) two times at 12-hour intervals. The following day, mice were injected intratumorally with oHSV119. For CPA (Bristol-Myers Squibb Co.,

157

Princeton, NJ) administration, drug was administered intraperitoneally (300mg/kg) 24 hours prior to viral injection390. The U87dEGFR cells were allowed to grow for 9 days and animals were subsequently randomly divided into groups that were injected intratumorally with either rQnestin34.5 in 3µl of HBSS or vehicle.

Flow cytometric analysis

Mononuclear cells were isolated from oHSV-infected brains using a previously described procedure with minor modifications375. In brief, 6,24, or 72 hours after infection, mice were sacrificed, brain tissue was harvested, and the tumor bearing hemispheres was placed in DMEM. The tissue was homogenized through a 70 µm strainer. Single cell preparations were re-suspended in 30% Percoll, overlaid on 70% Percoll (GE Healthcare,

Uppsala, Sweden) and centrifuged at 1300 x g for 30 minutes at 4oC. Cells at the 70%-

30% interface were collected, and washed with PBS. Cells isolated from the brain were treated with Fc Block (BD, San Jose, CA). Cells were then stained with anti-mouse immune cell surface markers for 30 min at 4oC. The following anti-mouse antibodies were used: CD3-FITC, DX5-PE, CD3-PercP, CD62L-APC, CD11b-PE (BD); CD62L-

FITC, CD69-FITC, Ly49d-APC, NKp46-FITC, CD11b-PercP, CD3-APC, CD45-APC,

(eBioscience, San Diego, CA); Ly-6c-FITC (Biolegend); and DX5-APC (Miltenyi

Biotec, Auburn, CA). Following antibody staining, cells were re-suspended in 1% formalin and analyzed using a FACS Calibur (Becton Dickinson, Mountain View, CA).

158

Quantitative real-time reverse transcriptase PCR

Total RNA from tumor bearing hemispheres or enriched human NK cells was isolated using the RNeasy Lipid Tissue Midi kit or RNeasy Mini kit, respectively (Qiagen,

Valencia, CA). A total of 5 µg of total RNA was reverse transcribed using random hexamers and the SuperScript First-Strand cDNA Synthesis System (Invitrogen).

Quantitative real-time PCR was done with cDNA samples diluted 1:100 in water and performed using SYBR Green PCR Master Mix and an ABI PRISM 7500 sequence detection system (Applied Biosystems, Foster City, CA). Murine primers used were as follows: 5’-TGGAGGAACTGGCAAAAGGA-3’ and 5’-

TGTTGCTGATGGCCTGATTG-3’ for IFN-γ; and 5’-

AAATGGTGAAGGTCGGTGTG-3’ and 5’- TGAAGGGGTCGTTGATGG-3’ for

GAPDH internal control.

NK cell isolation

NK cells were enriched from peripheral blood leukopacks of healthy donors (American

Red Cross, Columbus, Ohio) using RosetteSep cocktail (StemCell Technologies,

Vancouver, Canada). The enriched NK cells were then further purified using positive selection CD56 magnetic bead sorting (Miltenyi Biotec). Once NK cells were isolated, they were cultured overnight in 10% RPMI prior to being used in downstream applications.

159

Cytotoxicity assay

The panel of human glioblastoma cells was plated overnight at 104 cells/well. When noted, cells were treated with 5mM VPA for 14 hours prior to infection, washed, and then infected with rQnestin34.5119. Cells were infected for 8 hours with rQnestin34.5 (MOI

1.0) or mock infected. Human NK cells were added at an effector:target ratio of 12.5:1 in the presence of human IL-15 (10 ng/ml) (Miltenyi Biotec). The co-culture was allowed to proceed for 4 hours at 37oC. Glioblastoma lysis was assessed by measuring glucose 6- phosphate dehydrogenase released from lysed cells using the Vybrant Cytotoxicity Assay

Kit (Molecular Probes, Eugene, OR)402.

ELISA

Enriched human NK cells were cultured overnight in vitro in the presence or absence of 5 mM VPA. Additionally, NK cells were either untreated or stimulated overnight with various cytokines (IL-12, IL-15, IL-18 each at 10 ng/ml). Supernatants were collected for detection of IFN-γ by enzyme-linked immunosorbent assay (ELISA) as previously described110. Remaining cells were collected and processed for quantitative real-time reverse transcriptase PCR as described above. IL-12 was kindly provided by Genetics

Institute Inc. (Cambridge, Massachusetts), IL-15 was kindly provided by Amgen, and IL-

18 was purchased from R & D systems.

160

Western blotting

To assess STAT5, T-bet, and granzyme B signaling, direct cell lysates or total protein extractions were prepared from enriched human NK cells treated overnight in the absence or presence of varying amounts of VPA or cytokine stimulants. Western blotting was

110 performed as previously described . Assessment of actin by Western blotting was included to control for protein loading. Antibodies used for Western blotting are as follows: mouse mAb T-bet (Santa Cruz); rabbit mAb phospho-STAT5Tyr694 and rabbit mAb granzyme B (Cell Signaling, Danvers, MA).

Results

Valproic acid suppresses immune cell infiltration following oHSV administration

We confirmed the nature of immune cell infiltration after intracranially administering rQnestin34.5 into an orthotopic human glioblastoma (U87dEGFR) xenograft. At 6, 24, and 72 hours post infection, FACS analysis was used to quantify the presence of NK cells

(DX5+CD3-), microglia (CD45lowCD11b+), macrophages (CD45highCD11b+), and lymphocytes (CD45highCD11blow) into the tumor bearing hemisphere. Compared to mice treated with vehicle, there was a robust, time dependent increase in both the percentage and total numbers of recruited NK cells (Figure 5.1), macrophages (Figure 5.2), and lymphocytes (Figure 5.2) into tumor bearing hemispheres. As a control, prior to OV administration we pretreated mice with cyclophosphamide, a pan-immunosupressant that has been demonstrated by multiple groups to enhance OV efficacy through its immunomodulatory properties90,300,437. Similar to previous work form our laboratory90,

161

CPA robustly attenuated the recruitment of NK cells and macrophages within the tumor bearing hemisphere (Figure 5.1-2). Valproic acid has been shown to enhance OV efficacy through the inhibition of IFN-I and its downstream pathways119; degradation of cyclin D1 and VEGF inhibition323; and increased mitochondrial depolymerization and caspase 3/9 cleavage438. When administered prior to oHSV injection, we detected a significant decrease in the recruitment of NK cells and macrophages within the brain 6 and 24 hours after infection; however, the number of these cells in treated brains 72 hours after infection was not suppressed (Figure 5.1-2). Interestingly, the magnitude and time course of this suppression was not as robust compared when oHSV was paired with CPA, suggesting that VPA imparts a different profile of immune cell kinetics to the site of infection. These findings suggest that in addition to its ability to modulate intracellular antiviral mechanisms within the infected tumor, VPA attenuates the kinetics and abundance of recruited antiviral mediators initially after oHSV infection.

oHSV inoculation with VPA reduces the inflammatory response

While VPA attenuated the recruitment of NK cells and macrophages to the site of oHSV infection, we were interested in evaluating the phenotypic features of these cells. Similar to our findings in chapter 3, we found that oHSV administration induced a unique NK and macrophage phenotype for cells recruited to the site of infection. However, when

VPA was co-administered with oHSV, there was a decrease at 6 and 24 hours post- infection in surface expression of the early activation marker CD69377, the lymphocyte homing antigen CD62L378, the natural cytotoxicity receptor NKp46405, and the activating

162 receptors Ly49d380 (Figure 5.3). Additionally, macrophage staining of Ly-6c was significantly reduced with VPA co-therapy (Figure 5.3). In order to evaluate the consequences of these changes in activation state, we also found that at 24 hours post- infection, VPA suppressed IFN-γ expression following oHSV inoculation (Figure 5.4).

Collectively, these findings demonstrate that VPA not only suppresses the recruitment of these immune mediators, but also attenuates their activation in response to oncolytic viral infection.

VPA suppresses IFN-γ production in a Stat5/T-bet dependent fashion

To extend these in vivo findings, we used human NK cells to investigate the ability of

VPA treatment to suppress their cytotoxicity in vitro. NK cells are primed for IFN-γ production following exposure to IL-12, IL-15, and IL-18439. We confirmed this by evaluating IFN-γ transcript following overnight culture of NK cells from human donors in the presence of IL-12+IL-15 or IL-12+IL-18 (Figure 5.5). However, when VPA was added to the overnight culture, IFN-γ expression was significantly decreased. This suppression was also seen at the protein level. VPA abrogated secreted IFN-γ production in both the NK-92 cell line and across three separate human donors (Figure 5.6). Lastly, we found that VPA treatment inhibits monokine-stimulated IFN-γ production through attenuation of STAT5 phosphorylation and T-bet expression, a master regulator of IFN-γ gene expression (Figure 5.7).

163

Mediators of NK cell cytotoxicity are reduced by VPA

In addition to IFN-γ production, NK cell activation is also associated with cell-mediated cytotoxicity. In order to test the role of VPA in this process, we analyzed transcripts of perforin, granzyme B, granzyme A, and Fas ligand following overnight culture with increasing amounts of VPA across three donors. VPA mediated a striking dose- dependent attenuation in gene expression across all four mediators of NK cell cytotoxicity (Figure 5.8). This dose dependent reduction in granzyme B transcript also correlated with a decrease in protein levels (Figure 5.9). These findings demonstrate that VPA potently suppresses mechanisms of both NK cell mediated cytotoxicity and cytokine production.

NK cell mediated killing of oHSV infected glioblastoma is suppressed by VPA

Although VPA potently suppresses NK activation, we also tested its ability to suppress

NK mediated killing of OV infected glioblastoma. IL-15 activated human NK cells readily killed U251 and Gli36dEGFR glioblastoma cells that were first infected with oHSV (Fig 5.10a). When NK cells were first cultured overnight with either 0.3mM or

5.0 mM VPA, NK mediated killing was significantly suppressed. Previous reports have demonstrated that treatment of various tumor target cells increases their susceptibility for

NK mediated killing440-443. As a result, we investigated whether VPA exposure to glioblastoma prior to co-culture with NK cells would suppress killing. In our model, we found that pre-treating U251 or Gli36dEGFR cells did not enhance their susceptibility for

NK mediated killing; moreover, when both NK cells and glioblastoma were treated

164 overnight with VPA prior to co-culture, glioblastoma lysis was still suppressed. We also extended our finding that VPA suppresses NK mediated killing across two additional donors when cultured with the U87dEGR glioblastoma cell line and the X12 “stem-like” glioblastoma cells (Fig 5.10b).

Discussion

Combining oncolytic viruses with pharmacological modulators has proven in preclinical models to be a promising method for improving OV therapy. As these agents gradually move towards testing in clinical trials, it is important to gain a greater appreciation for their mechanism of action. For instance, CPA has been shown to enhance OV efficacy through circumventing not only complement mediated viral depletion but also through the attenuation of the antiviral response mediated by early immune responders90,118.

VPA has been more recently described as an effective adjuvant to oHSV. Consequently, its multiple mechanisms of action have not been fully investigated. Apart from its anticancer properties, the combination treatment of VPA and oHSV achieved success through VPA’s inhibition of IFN-β and IFN-mediated proteins STAT1, PKR and PML in infected cells119. Trichostatin A (TSA), a similar HDACi, has also been paired with oHSV to increase oncolysis. In the context of TSA, therapeutic efficacy was achieved through inhibition of cyclin D1 and its concomitant arrest in cell cycle progression320-322.

Antitumor efficacy has also been achieved by an increase in TSA mediated mitochondrial depolymerization and cleavage of caspases 3 and 9 when paired with oncolytic VSV185.

165

Collectively, these findings demonstrate that HDAC inhibition is facilitating successful

OV therapy through various mechanisms within infected cells. However, the pleiotropic nature of these agents has not been studied in the context of immune cells recruited to the site of infection.

Our novel findings focus on VPA mediated suppression of immune cell recruitment and activation following oHSV infection. Additionally, we have studied the ability of VPA to suppress NK cell activation and cytotoxicity in an in vitro model. The in vivo effect of

VPA demonstrates that the recruitment of innate immune cells is suppressed; thereby, creating an environment that is more conducive to initial viral replication90. Interestingly, the magnitude of this response is shorter in nature than what is seen with CPA co- therapy. As result, VPA provides an interesting alternative co-therapy for oHSV. Rather than achieving profound attenuation of inflammatory mediators at all timepoints considered, VPA pre-treatment results in reduced inflammatory cell recruitment at initial timepoints after infection with more significant recruitment occurring at later timepoints.

Future studies will be needed to elucidate the relevancy of these findings when comparing CPA and VPA co-therapy in the clinical setting. In particular, attention should be given to differences in dosing regimens, toxicity associated immunosuppression, and the ability to bypass initial antiviral mediators while eliciting downstream antitumor immunity.

166

Our in vitro findings that VPA significantly suppressed NK cell production of both IFN-γ in a Stat5/T-bet dependent fashion and mediators of cell mediated cytotoxicity are the first reports of this kind and highlight interesting dichotomies within the field of NK biology. In accordance with our results, there is one report by Ogbomo et al. that found pretreatment of NK cells with the HDACi suberoylanilide hydroxamic acid or valproic acid significantly inhibited NK cell cytotoxicity of human leukemic cells444. This was associated with reduced NKp30 and NKp46 natural cytotoxicity receptor expression.

Interestingly, these findings corresponded with our in vivo findings that VPA co-therapy reduced the expression of NKp46 on recruited NK cells. Lastly, the authors determined that VPA induced suppression of NK cell cytotoxicity occurred through impaired granule exocytosis and inhibition of NF-kB activation444.

In contrast to our findings and the work of Ogbomo et al, several groups have reported that HDACi can enhance NK mediated killing. Zhang et al. have reported that enhanced susceptibility to NK mediated killing occurs following sodium butyrate treatment of

HeLA and HepG2 tumor cells lines due to enhanced expression of the NKG2D ligands

MHC class I-related chain molecules A and B (MICA and MICB)440. Skov et al. similarly found that HDAC inhibition led to increases in both expression of MICA/B on a variety of tumors cells and subsequent NK mediated killing442. Lastly, Armeanu et al. detected increased MICA/MICB expression in hepatocellular carcinoma cells treated with VPA, rendering these cells more susceptible to NK mediated lysis443.

167

In order to understand these differences, it essential to differentiate between the effect of

HDACi on NK cells versus target cells. A corpus of work has demonstrated that HDACi exposure to tumor cells results in MICA/B induction and downstream susceptibility to

NK cells. However, these studies did not evaluate NK cells exposed to HDACi. Based on our findings and the work of Ogbomo et al., it is likely that HDACi results in suppression of NK cell cytotoxicity while simultaneously resulting in increased

MICA/MICB expression on target cells. However, our study is the first to report that treatment of NK cells with VPA resulted in reduced NK mediated killing regardless of

VPA exposure to the glioblastoma target cells. While this does not argue against using

VPA to induce enhanced NK mediated killing, it does call for caution when designing immunotherapy and adoptive therapy trials to ensure that the HDACi-induced increase in

NK ligand expression is not counteracted by suppression of NK cytotoxicity.

As oncolytic viral co-therapies with pharmacological agents are increasingly pursued for implementation in the clinic, it is essential to understand the mechanisms through which efficacy is being achieved. In this study, VPA co-administration was shown to attenuate the elicited host immune response to oncolytc virus. This is in addition to the previously uncovered mechanism of circumventing host-cell antiviral defenses. In addition, we uncover novel mechanism by which VPA suppresses human NK cell cytotoxicity and

IFN-γ production. Taken together, these findings provide insights that can be used to enhance subsequent studies that attempt to target HDAC inhibition and suppress NK cell activity.

168

A. continued

Figure 5.1: VPA attenuates NK recruitment initially after oHSV infection.

(A) Human U87dEGFR cells (105) were implanted intracranially into athymic mice brains (n=3/group) and allowed to grow for 9 days. In addition to vehicle treated mice, the following IP drug injections were performed: CPA 48 hours prior to oHSV administration and VPA 12 and 24 hours prior to oHSV administration. Following drug treatment, rQnestin34.5 (104 pfu/3µl vehicle) or vehicle was then stereotactically inoculated, using the same coordinates. Tumor bearing hemispheres were harvested 6,

24, or 72 hours after infection (n=4-5/group) so that the percentage of NK cells in tumor bearing hemispheres could be quantified using FACS. (B) Athymic mice were implanted with tumor, treated, and sacrificed as described in (A). The total number of infiltrating

NK cells in tumor bearing hemispheres could be quantified using FACS. Error bars represent +/- standard deviation.

169

Figure 5.1 continued

B.

170

Figure 5.2: VPA attenuates macrophage recruitment initially after oHSV infection.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice and allowed to grow for 9 days. In addition to vehicle treated mice, the following IP drug injections were performed: CPA 48 hours prior to oHSV administration and VPA 12 and

24 hours prior to oHSV administration. Following drug treatment, rQnestin34.5 (104 pfu/3µl vehicle) or vehicle was stereotactically inoculated, using the same coordinates.

Tumor bearing hemispheres were harvested 6, 24, or 72 hours after infection (n=4-

5/group) in order to quantify by FACS the percentage microglia (CD45lowCD11b+), macrophage (CD45highCD11b+), and lymphocyte (CD45highCD11blow) cell populations in tumor bearing hemispheres following treatment (n=4-6 mice/group).

171

Figure 5.3: VPA suppresses NK activation initially after oHSV infection.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice and allowed to grow for 9 days. Mice were treated with either vehicle or VPA 12 and 24 hours prior to oHSV administration. Following drug treatment, rQnestin34.5 (104 pfu/3µl vehicle) or vehicle was stereotactically inoculated, using the same coordinates.

Six or 24 hours later, mice were sacrificed and tumor bearing hemispheres were processed so that the NK cells and macrophages could be analyzed using FACS.

172

Figure 5.4: VPA attenuates IFN-γ expression following oncolytic viral infection.

Human U87dEGFR tumors (105 cells) were implanted intracranially into athymic mice

(n=4-5/group) and allowed to grow for 9 days. Mice were treated with either vehicle or

VPA 12 and 24 hours prior to oHSV administration. Following drug treatment, rQnestin34.5 (104 pfu/3µl vehicle) or vehicle was stereotactically inoculated, using the same coordinates. Mice were sacrificed 24 hours later for mRNA isolation of the tumor bearing hemisphere. Following mRNA conversion to cDNA, the expression of IFN-γ was evaluated with RT-qPCR. * P < 0.05; *** P < 0.001. Error bars represent +/- standard deviation.

173

Figure 5.5: VPA reduces human NK cell expression of IFN-γ transcript.

Enriched human NK cells from multiple donors were cultured overnight with varying combinations of IL-12 and either IL-15 or IL-18 (each cytokine used at 100 ng/mL) in the absence or presence of 5.0 mM VPA. The following day, the NK cells were collected for mRNA isolation. Following mRNA conversion to cDNA, the expression of IFN-γ was evaluated with RT-qPCR. Error bars represent +/- standard deviation.

174

A. continued

Figure 5.6: VPA attenuates cytokine mediated induction of NK cell IFN-γ secretion.

(A) The human NK cell line, NK-92, was cultured overnight with IL-12, IL-15 IL-18, or

IL-12/IL-15 (each cytokine used at 100 ng/mL) in the absence or presence of 5.0 mM

VPA. The following day, supernatant was collected and analyzed using an IFN-γ ELISA assay. (B) Enriched human NK cells from multiple donors were cultured overnight with varying combinations of IL-12 and either IL-15 or IL-18 (each cytokine used at 100 ng/mL) in the absence or presence of 5.0 mM VPA. The following day, supernatant was collected and analyzed using an IFN-γ ELISA assay. Error bars represent +/- standard deviation.

175

Figure 5.6 continued

B.

176

Figure 5.7: VPA antagonizes cytokine induced Stat5 and Tbet upregulation.

Enriched human NK cells were cultured overnight with IL-12, IL-15 IL-18, or IL-12 and

IL-15/IL-18 (each cytokine used at 100 ng/mL) in the absence or presence of 5.0 mM

VPA. The following day, cells were collected, lysed, and cell lysate was analyzed using

Western blot for Stat5 and Tbet. Actin was used as a loading control.

177

Figure 5.8: VPA suppresses NK cell cytotoxicity genes in a dose dependent manner.

Enriched human NK cells from multiple donors were cultured overnight either in the absence or increasing amounts of VPA. The following day, the NK cells were collected for mRNA isolation. Following mRNA conversion to cDNA, the expression of Perforin,

Fas ligand, granzyme B, and granzyme A was evaluated with RT-qPCR. Error bars represent +/- standard deviation.

178

Figure 5.9: VPA reduces granzyme B protein levels in a dose dependent manner.

Enriched human NK cells were cultured overnight either in the absence or increasing amounts of VPA. The following day, cells were collected, lysed, and cell lysate was analyzed using Western blot for granzyme B. Actin was used as a loading control.

179

A. continued

Figure 5.10: NK cell cytotoxicity is suppressed following exposure to VPA.

Enriched human NK cells were cultured overnight either in the absence or presence of varying amounts of VPA (0.3 or 5.0 mM). Gli36dEGFR and U251 cells lines were plated (104 cells/well) and allowed to grow overnight in the absence or presence of 5.0 mM VPA. Twelve hours after the addition of VPA to the glioblastoma cells, the cells were washed, fresh media was added, and infected with rQnestin34.5 (MOI 1.0) or mock- infected for 8 hours. The mock or VPA treated NK cells were subsequently added to the glioblastoma co-culture at an effector:target ratio of 12.5:1 in the presence of IL-15. Four hours later glioblastoma lysis was assessed. (B) Enriched human NK cells from multiple donors were cultured as described in (A). The U87dEGFR human glioblastoma cell line or primary human glioblastoma cells enriched for stem-like cell properties (X12) were plated (104 cells/well) and infected as described in (A). The mock or VPA treated NK cells were subsequently added to the glioblastoma co-culture at an effector: target ratio of

12.5:1 in the presence of IL-15. Four hours later glioblastoma lysis was assessed. Error bars represent +/- standard deviation. 180

Figure 5.10 continued

B.

181

Chapter 6: Future directions

Nearly two decades after the first published report of oncolytic viral therapy38, investigators using these viruses have made remarkable progress in their preclinical testing and evaluation in clinical trials. Moreover, the recent approval of the H101 oncolytic adenovirus in China170 combined with numerous clinical trials in place within the United States and Europe suggests that virotherapy will gradually enter the armamentarium of tomorrow’s physicians. In order to achieve widespread clinical applicability, however, certain obstacles must be overcome. For instance, infected cells have various antiviral defense mechanisms that must be circumvented to achieve sustained viral replication; however, circumventing this response must be countered with concerns about uncontrolled viral replication and toxicity. Additionally, host immunity, particularly innate immunity, is a first line of defense against foreign pathogens that has been demonstrated to impede virotherapy; however, immune suppression potentially impedes the antitumor immune response that has been shown to synergize with viral oncolysis. Lastly, co-administration of pharmacological agents that cooperate with viral mediated tumor clearance shows significant promise; however, the comparative effectiveness of treating various tumors with the appropriate virus/drug combination must be thoroughly evaluated in clinical trials.

182

In recent years, significant attention has been directed towards the host immune response to oncolytic viruses. In particular, the role of initial immune responder cells, including

NK cells and macrophages, has been questioned. With their antiviral and antitumor properties combined with their ability to mediate macrophage activation, the NK cell response to virotherapy has elicited significant attention. In the context of oncolytic HSV treatment for glioblastoma, we find that NK cells are activated and rapidly recruited to the site of infection; are deleterious to viral mediated tumor clearance; mediate macrophage and microglia activation; and preferentially kill virally infected cells through their NKp30 and NKp46 receptors. Additionally, VPA’s combination with oHSV alters the kinetics of immune cell recruitment following infection; suppresses NK mediated cytotoxicity; and attenuates IFN-γ production through suppression of Stat5 and T-bet.

Despite these findings, significant work is needed to extend these results. While murine xenograft models allow for investigators to study the clearance of human tumors by oncolytc viruses, this model prevents the evaluation of how innate immunity coordinates with adaptive immunity following viral infection. In the context of oHSV therapy for glioblastoma, this raises a significant challenge. While there are a number of orthotopic glioblastoma models, the majority of them use strains of mice, in particular C57BL/6, that are naturally resistant to HSV infection399,445. In order to extend the significance of our findings and more accurately model what is happening in humans, it is critical that future work develops a fully syngeneic, orthotopic model of glioblastoma that is

183 susceptible to in vivo oHSV infection and replication. Taken further, numerous reports have highlighted the histological heterogeneity of glioblastoma8,125 and their reliance on tumor initiating cells446. As reliable immunocompetent animal models are identified, it will also be important to study the role of NK cells in response to oHSV treatment of various histologic subtypes of glioblastoma (e.g. pro-neural, mesenchymal, etc.) and glioblastoma initiating cells.

In order to fully investigate the NK mediated mechanisms that limit oHSV efficacy, future studies should also attempt to test oHSV in a syngeneic mouse model with specific

NK deficiencies. For instance, it would be worthwhile to compare OV efficacy in mice possessing NK cells, lacking NK cells, with NK cells deficient in INF-γ production, with

NK cells deficient in cell mediated cytotoxicity, and with NK cells deficient in NKp46.

By testing for OV efficacy and downstream immune cell activation, including macrophage and T cell polarization, in each of these groups, it would be possible to delineate the critical NK components in antiviral immunity to oHSV. Additionally,

NKp46 is the only NCR present on murine NK cells. As a result, we are limited in our ability to test the significance of NKp44 and NKp30 against oHSV in vivo. To circumvent this problem, it would be possible to evaluate oHSV in the context of a humanized mouse model447 where NKp46, NKp44, and NKp30 are expressed.

While a variety of co-therapies have been shown co-operate with viral oncolysis through immune cell suppression, additional work can be done to fine-tune this approach. For

184 instance, it is shortsighted to think that anti-tumor immunity has no part in viral oncolysis. Rather, future work must identify the delicate balance between an initial suppression of antiviral immunity thereby facilitating initial rounds of viral replication followed by the stimulation of anti-tumor immunity against tumor or viral antigens. This could potentially take the form of targeting NK cell depletion in a temporal manner whereby NK cells are attenuated initially after viral infection and then allowed to repopulate the site of infection a few days after viral inoculation. This later response would take advantage of their anti-tumor properties, macrophage activating properties, and their ability to induce a Th1 response.

An additional approach could focus on M1/M2 macrophage polarization following oHSV infection. Clinical reports have confirmed that human glioblastoma is typically associated with generalized immunosuppression, TGF-β production, and an M2 macrophage phenotype448. Previous work from our laboratory has demonstrated that macrophage activation in the context of oHSV inoculation is detrimental to OV efficacy86. In chapter 3, we also show that oHSV inoculation results in a robust inflammatory response within 72 hours after infection that is associated with an M1 macrophage inflammatory state. Taken together, future studies could attempt to discern whether a temporary maintenance of the M2 phenotype initially after infection followed by a switch towards an inflammatory M1 state is effective at initially enhancing viral titers while eliciting anti-tumor immunity at later timepoints.

185

A particularly novel discovery from chapter 4 was the role of NKp30/NKp46 in the clearance of glioblastoma infected with oHSV. Not only were these receptors involved in this response but also the ligands for these receptors were robustly expressed within virally infected glioblastoma. In order to expand the relevancy of these findings, future work will need to investigate the identity of these ligands in the GFPhigh population; whether the ligands are viral in origin or expressed following cellular stress; and whether they are the same ligands that are expressed following the exposure of glioblastoma to

TMZ. The identify of ligands expressed after oHSV infection will be useful on multiple fronts. First, it could potentially be targeted to enhance viral efficacy. If NKp30/NKp46 are the key receptors mediating premature viral clearance, suppressing the expression of these ligands could be particularly advantageous. Targeted suppression could be achieved through either pharmacologic means or through the creation of a novel oHSV that expresses a decoy for NKp30/NKp46 or inhibits ligand presentation on the infected cell surface. Second, in chapter 4 we demonstrate that NK cells preferentially kill glioblastoma cells infected with oHSV via NKp30/NKp46. While this is deleterious to initial infection and replication of oHSV, it represents a novel target for mediating anti- tumor immunity. As a result, in instances where NK mediated anti-tumor immunity is deemed beneficial, such as later time points after infection once productive viral replication is established, eliciting NKp30/NKp46 mediated tumor killing could be pursued as a viable therapeutic option.

186

More broadly, specific oHSV mutants could be generated that are aimed at suppressing generalized NK activation following oHSV infection. For instance, HSV encoded ICP0 and ICP27 regulate Stat1 activation and IFN-I expression258,449-451. Additionally, UL41,

US11, and US3 contribute to the inhibition of the host IFN response452. Meanwhile, data from Huard et al. suggest that ICP47 is an important mediator in NK cell activation453.

Taken collectively, future work could focus on constructing viral mutants of HSV genes involved in NK activation and then testing their efficacy in vivo and in vitro. By evaluating such a panel of mutant oHSVs, a novel construct could be identified that suppresses NK cell activation and viral clearance.

With a number of oncolytic viral clinical trials in the pipeline, it will be critical for investigators to include the evaluation of NK cells in the immunological response to viral administration. Based on our preclinical work, attention should be directed towards NK cell numbers in the periphery, within the tumor microenvironment, and their distribution within the virally infected tumor. Moreover, the activation/developmental state of these

NK cells should be evaluated. For instance, NK cells should be tested for CD56 and

CD94 expression, whether the recruited/circulating NK cells are NCRbright or NCRdim, and their functional capacity. By collecting this data from human samples, we can test the validity of our preclinical models and guide future experimental trials.

187

References

1. Gorlia, T., et al. Nomograms for predicting survival of patients with newly diagnosed glioblastoma: prognostic factor analysis of EORTC and NCIC trial 26981-22981/CE.3. Lancet Oncol 9, 29-38 (2008). 2. Maris, J.M., Hogarty, M.D., Bagatell, R. & Cohn, S.L. Neuroblastoma. Lancet 369, 2106-2120 (2007). 3. Westphal, M., et al. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma. Neuro Oncol 5, 79-88 (2003). 4. Stupp, R., et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352, 987-996 (2005). 5. Stupp, R., et al. Changing paradigms--an update on the multidisciplinary management of malignant glioma. Oncologist 11, 165-180 (2006). 6. Walker, M.D., et al. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas. A cooperative clinical trial. J Neurosurg 49, 333-343 (1978). 7. Hochberg, F.H. & Pruitt, A. Assumptions in the radiotherapy of glioblastoma. Neurology 30, 907-911 (1980). 8. Wen, P.Y. & Kesari, S. Malignant gliomas in adults. N Engl J Med 359, 492-507 (2008). 9. Chiocca, E.A. Oncolytic viruses. Nat Rev Cancer 2, 938-950 (2002). 10. Aghi, M., Visted, T., Depinho, R.A. & Chiocca, E.A. Oncolytic herpes virus with defective ICP6 specifically replicates in quiescent cells with homozygous genetic mutations in p16. Oncogene 27, 4249-4254 (2008). 11. Nakano, K., et al. Herpes simplex virus targeting to the EGF receptor by a gD- specific soluble bridging molecule. Mol Ther 11, 617-626 (2005). 12. Hallak, L.K., Merchan, J.R., Storgard, C.M., Loftus, J.C. & Russell, S.J. Targeted measles virus vector displaying echistatin infects endothelial cells via alpha(v)beta3 and leads to tumor regression. Cancer Res 65, 5292-5300 (2005). 13. Hedley, S.J., et al. An adenovirus vector with a chimeric fiber incorporating stabilized single chain antibody achieves targeted gene delivery. Gene Ther 13, 88-94 (2006). 14. Borovjagin, A.V., et al. Complex mosaicism is a novel approach to infectivity enhancement of adenovirus type 5-based vectors. Cancer Gene Ther 12, 475-486 (2005).

188

15. Balachandran, S. & Barber, G.N. Defective translational control facilitates vesicular stomatitis virus oncolysis. Cancer Cell 5, 51-65 (2004). 16. Balachandran, S. & Barber, G.N. Vesicular stomatitis virus (VSV) therapy of tumors. IUBMB Life 50, 135-138 (2000). 17. Shinozaki, K., Ebert, O., Kournioti, C., Tai, Y.S. & Woo, S.L. Oncolysis of multifocal hepatocellular carcinoma in the rat liver by hepatic artery infusion of vesicular stomatitis virus. Mol Ther 9, 368-376 (2004). 18. Shinozaki, K., Ebert, O., Suriawinata, A., Thung, S.N. & Woo, S.L. Prophylactic alpha interferon treatment increases the therapeutic index of oncolytic vesicular stomatitis virus virotherapy for advanced hepatocellular carcinoma in immune- competent rats. J Virol 79, 13705-13713 (2005). 19. Cesaire, R., et al. Oncolytic activity of vesicular stomatitis virus in primary adult T-cell leukemia. Oncogene 25, 349-358 (2006). 20. Stojdl, D.F., et al. Exploiting tumor-specific defects in the interferon pathway with a previously unknown oncolytic virus. Nat Med 6, 821-825 (2000). 21. Anderson, B.D., Nakamura, T., Russell, S.J. & Peng, K.W. High CD46 receptor density determines preferential killing of tumor cells by oncolytic measles virus. Cancer Res 64, 4919-4926 (2004). 22. Peng, K.W., et al. Biodistribution of oncolytic measles virus after intraperitoneal administration into Ifnar-CD46Ge transgenic mice. Hum Gene Ther 14, 1565- 1577 (2003). 23. Russell, S.J. RNA viruses as virotherapy agents. Cancer Gene Ther 9, 961-966 (2002). 24. Coffey, M.C., Strong, J.E., Forsyth, P.A. & Lee, P.W. Reovirus therapy of tumors with activated Ras pathway. Science 282, 1332-1334 (1998). 25. Yang, W.Q., et al. Reovirus prolongs survival and reduces the frequency of spinal and leptomeningeal metastases from medulloblastoma. Cancer Res 63, 3162-3172 (2003). 26. Yang, W.Q., et al. Reovirus as an experimental therapeutic for brain and leptomeningeal metastases from breast cancer. Gene Ther 11, 1579-1589 (2004). 27. Tseng, J.C., et al. In vivo antitumor activity of Sindbis viral vectors. J Natl Cancer Inst 94, 1790-1802 (2002). 28. Tseng, J.C., et al. Systemic tumor targeting and killing by Sindbis viral vectors. Nat Biotechnol 22, 70-77 (2004). 29. Ochiai, H., et al. Targeted therapy for glioblastoma multiforme neoplastic meningitis with intrathecal delivery of an oncolytic recombinant poliovirus. Clin Cancer Res 12, 1349-1354 (2006). 30. Merrill, M.K., et al. Poliovirus receptor CD155-targeted oncolysis of glioma. Neuro Oncol 6, 208-217 (2004). 31. Solecki, D.J., Gromeier, M., Mueller, S., Bernhardt, G. & Wimmer, E. Expression of the human poliovirus receptor/CD155 gene is activated by sonic hedgehog. J Biol Chem 277, 25697-25702 (2002).

189

32. Gromeier, M., Lachmann, S., Rosenfeld, M.R., Gutin, P.H. & Wimmer, E. Intergeneric poliovirus recombinants for the treatment of malignant glioma. Proc Natl Acad Sci U S A 97, 6803-6808 (2000). 33. Lorence, R.M., et al. Complete regression of human fibrosarcoma xenografts after local Newcastle disease virus therapy. Cancer Res 54, 6017-6021 (1994). 34. Lorence, R.M., et al. Complete regression of human neuroblastoma xenografts in athymic mice after local Newcastle disease virus therapy. J Natl Cancer Inst 86, 1228-1233 (1994). 35. Lorence, R.M., et al. Overview of phase I studies of intravenous administration of PV701, an oncolytic virus. Curr Opin Mol Ther 5, 618-624 (2003). 36. Tai, C.K., Wang, W.J., Chen, T.C. & Kasahara, N. Single-shot, multicycle suicide gene therapy by replication-competent retrovirus vectors achieves long-term survival benefit in experimental glioma. Mol Ther 12, 842-851 (2005). 37. Wang, W., et al. Use of replication-competent retroviral vectors in an immunocompetent intracranial glioma model. Neurosurg Focus 20, E25 (2006). 38. Martuza, R.L., Malick, A., Markert, J.M., Ruffner, K.L. & Coen, D.M. Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 252, 854-856 (1991). 39. Mineta, T., Rabkin, S.D., Yazaki, T., Hunter, W.D. & Martuza, R.L. Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nat Med 1, 938-943 (1995). 40. Advani, S.J., et al. Enhancement of replication of genetically engineered herpes simplex viruses by ionizing radiation: a new paradigm for destruction of therapeutically intractable tumors. Gene Ther 5, 160-165 (1998). 41. Advani, S.J., et al. Replication-competent, nonneuroinvasive genetically engineered herpes virus is highly effective in the treatment of therapy-resistant experimental human tumors. Cancer Res 59, 2055-2058 (1999). 42. Boviatsis, E.J., et al. Long-term survival of rats harboring brain neoplasms treated with ganciclovir and a herpes simplex virus vector that retains an intact thymidine kinase gene. Cancer Res 54, 5745-5751 (1994). 43. Boviatsis, E.J., et al. Antitumor activity and reporter gene transfer into rat brain neoplasms inoculated with herpes simplex virus vectors defective in thymidine kinase or ribonucleotide reductase. Gene Ther 1, 323-331 (1994). 44. Benencia, F., et al. HSV oncolytic therapy upregulates interferon-inducible chemokines and recruits immune effector cells in ovarian cancer. Mol Ther 12, 789-802 (2005). 45. Thomas, D.L. & Fraser, N.W. HSV-1 therapy of primary tumors reduces the number of metastases in an immune-competent model of metastatic breast cancer. Mol Ther 8, 543-551 (2003). 46. Kesari, S., et al. Therapy of experimental human brain tumors using a neuroattenuated herpes simplex virus mutant. Lab Invest 73, 636-648 (1995). 47. Miller, C.G. & Fraser, N.W. Role of the immune response during neuro- attenuated herpes simplex virus-mediated tumor destruction in a murine intracranial melanoma model. Cancer Res 60, 5714-5722 (2000). 190

48. Smith, K.D., et al. Activated MEK suppresses activation of PKR and enables efficient replication and in vivo oncolysis by Deltagamma(1)34.5 mutants of herpes simplex virus 1. J Virol 80, 1110-1120 (2006). 49. Curiel, D.T. & Rancourt, C. Conditionally replicative adenoviruses for cancer therapy. Adv Drug Deliv Rev 27, 67-81 (1997). 50. Curiel, D.T. The development of conditionally replicative adenoviruses for cancer therapy. Clin Cancer Res 6, 3395-3399 (2000). 51. Adachi, Y., et al. A midkine promoter-based conditionally replicative adenovirus for treatment of pediatric solid tumors and bone marrow tumor purging. Cancer Res 61, 7882-7888 (2001). 52. Hakkarainen, T., Hemminki, A., Curiel, D.T. & Wahlfors, J. A conditionally replicative adenovirus that codes for a TK-GFP fusion protein (Ad5Delta24TK- GFP) for evaluation of the potency of oncolytic virotherapy combined with molecular chemotherapy. Int J Mol Med 18, 751-759 (2006). 53. Hay, J.G. The potential impact of hypoxia on the success of oncolytic virotherapy. Curr Opin Mol Ther 7, 353-358 (2005). 54. Sauthoff, H., et al. Intratumoral spread of wild-type adenovirus is limited after local injection of human xenograft tumors: virus persists and spreads systemically at late time points. Hum Gene Ther 14, 425-433 (2003). 55. Kuppuswamy, M., et al. Oncolytic adenovirus that overproduces ADP and replicates selectively in tumors due to hTERT promoter-regulated E4 gene expression. Gene Ther 12, 1608-1617 (2005). 56. Toth, K., et al. An oncolytic adenovirus vector combining enhanced cell-to-cell spreading, mediated by the ADP cytolytic protein, with selective replication in cancer cells with deregulated wnt signaling. Cancer Res 64, 3638-3644 (2004). 57. Fueyo, J., et al. A mutant oncolytic adenovirus targeting the Rb pathway produces anti-glioma effect in vivo. Oncogene 19, 2-12 (2000). 58. Jiang, H., McCormick, F., Lang, F.F., Gomez-Manzano, C. & Fueyo, J. Oncolytic adenoviruses as antiglioma agents. Expert Rev Anticancer Ther 6, 697-708 (2006). 59. O'Shea, C.C., et al. Late viral RNA export, rather than p53 inactivation, determines ONYX-015 tumor selectivity. Cancer Cell 6, 611-623 (2004). 60. McCormick, F. Future prospects for oncolytic therapy. Oncogene 24, 7817-7819 (2005). 61. Kim, J.H., et al. Systemic armed oncolytic and immunologic therapy for cancer with JX-594, a targeted poxvirus expressing GM-CSF. Mol Ther 14, 361-370 (2006). 62. Timiryasova, T.M., et al. Antitumor effect of vaccinia virus in glioma model. Oncol Res 11, 133-144 (1999). 63. McCart, J.A., et al. Complex interactions between the replicating oncolytic effect and the enzyme/prodrug effect of vaccinia-mediated tumor regression. Gene Ther 7, 1217-1223 (2000).

191

64. McCart, J.A., et al. Oncolytic vaccinia virus expressing the human somatostatin receptor SSTR2: molecular imaging after systemic delivery using 111In- pentetreotide. Mol Ther 10, 553-561 (2004). 65. Guo, Z.S., et al. The enhanced tumor selectivity of an oncolytic vaccinia lacking the host range and antiapoptosis genes SPI-1 and SPI-2. Cancer Res 65, 9991- 9998 (2005). 66. Wang, G., et al. Infection of human cancer cells with myxoma virus requires Akt activation via interaction with a viral ankyrin-repeat host range factor. Proc Natl Acad Sci U S A 103, 4640-4645 (2006). 67. Lun, X., et al. Myxoma virus is a novel oncolytic virus with significant antitumor activity against experimental human gliomas. Cancer Res 65, 9982-9990 (2005). 68. Crompton, A.M. & Kirn, D.H. From ONYX-015 to armed vaccinia viruses: the education and evolution of oncolytic virus development. Curr Cancer Drug Targets 7, 133-139 (2007). 69. Chiocca, E.A., et al. A phase I open-label, dose-escalation, multi-institutional trial of injection with an E1B-Attenuated adenovirus, ONYX-015, into the peritumoral region of recurrent malignant gliomas, in the adjuvant setting. Mol Ther 10, 958- 966 (2004). 70. Fong, Y., et al. A herpes oncolytic virus can be delivered via the vasculature to produce biologic changes in human colorectal cancer. Mol Ther 17, 389-394 (2009). 71. Forsyth, P., et al. A phase I trial of intratumoral administration of reovirus in patients with histologically confirmed recurrent malignant gliomas. Mol Ther 16, 627-632 (2008). 72. Hu, J.C., et al. A phase I study of OncoVEXGM-CSF, a second-generation oncolytic herpes simplex virus expressing granulocyte macrophage colony- stimulating factor. Clin Cancer Res 12, 6737-6747 (2006). 73. Kaufman, H.L. & Bines, S.D. OPTIM trial: a Phase III trial of an oncolytic herpes virus encoding GM-CSF for unresectable stage III or IV melanoma. Future Oncol 6, 941-949. 74. Kaufman, H.L., et al. Local and distant immunity induced by intralesional vaccination with an oncolytic herpes virus encoding GM-CSF in patients with stage IIIc and IV melanoma. Ann Surg Oncol 17, 718-730. 75. Liu, T.C., Hwang, T., Park, B.H., Bell, J. & Kirn, D.H. The targeted oncolytic poxvirus JX-594 demonstrates antitumoral, antivascular, and anti-HBV activities in patients with hepatocellular carcinoma. Mol Ther 16, 1637-1642 (2008). 76. Small, E.J., et al. A phase I trial of intravenous CG7870, a replication-selective, prostate-specific antigen-targeted oncolytic adenovirus, for the treatment of hormone-refractory, metastatic prostate cancer. Mol Ther 14, 107-117 (2006). 77. Lawler, S.E., Peruzzi, P.P. & Chiocca, E.A. Genetic strategies for brain tumor therapy. Cancer Gene Ther 13, 225-233 (2006). 78. Varghese, S. & Rabkin, S.D. Oncolytic herpes simplex virus vectors for cancer virotherapy. Cancer Gene Ther 9, 967-978 (2002).

192

79. Markert, J.M., et al. Conditionally replicating herpes simplex virus mutant, G207 for the treatment of malignant glioma: results of a phase I trial. Gene Ther 7, 867- 874 (2000). 80. Markert, J.M., et al. Phase Ib trial of mutant herpes simplex virus G207 inoculated pre-and post-tumor resection for recurrent GBM. Mol Ther 17, 199- 207 (2009). 81. Rampling, R., et al. Toxicity evaluation of replication-competent herpes simplex virus (ICP 34.5 null mutant 1716) in patients with recurrent malignant glioma. Gene Ther 7, 859-866 (2000). 82. Papanastassiou, V., et al. The potential for efficacy of the modified (ICP 34.5(-)) herpes simplex virus HSV1716 following intratumoural injection into human malignant glioma: a proof of principle study. Gene Ther 9, 398-406 (2002). 83. Harrow, S., et al. HSV1716 injection into the brain adjacent to tumour following surgical resection of high-grade glioma: safety data and long-term survival. Gene Ther 11, 1648-1658 (2004). 84. Freeman, A.I., et al. Phase I/II trial of intravenous NDV-HUJ oncolytic virus in recurrent glioblastoma multiforme. Mol Ther 13, 221-228 (2006). 85. Forsyth, P., et al. A phase I trial of intratumoral administration of reovirus in patients with histologically confirmed recurrent malignant gliomas. Mol Ther 16, 627-632 (2008). 86. Fulci, G., et al. Depletion of peripheral macrophages and brain microglia increases brain tumor titers of oncolytic viruses. Cancer Res 67, 9398-9406 (2007). 87. Todo, T., et al. Systemic antitumor immunity in experimental brain tumor therapy using a multimutated, replication-competent herpes simplex virus. Hum Gene Ther 10, 2741-2755 (1999). 88. Toda, M., Rabkin, S.D., Kojima, H. & Martuza, R.L. Herpes simplex virus as an in situ cancer vaccine for the induction of specific anti-tumor immunity. Hum Gene Ther 10, 385-393 (1999). 89. Nakano, K., Todo, T., Chijiiwa, K. & Tanaka, M. Therapeutic efficacy of G207, a conditionally replicating herpes simplex virus type 1 mutant, for gallbladder carcinoma in immunocompetent hamsters. Mol Ther 3, 431-437 (2001). 90. Fulci, G., et al. Cyclophosphamide enhances glioma virotherapy by inhibiting innate immune responses. Proc Natl Acad Sci USA 103, 12873-12878 (2006). 91. Kurozumi, K., et al. Effect of Tumor Microenvironment Modulation on the Efficacy of Oncolytic Virus Therapy. J Natl Cancer Inst (2007). 92. Dehdashti, A.R., Hegi, M.E., Regli, L., Pica, A. & Stupp, R. New trends in the medical management of glioblastoma multiforme: the role of temozolomide chemotherapy. Neurosurg Focus 20, E6 (2006). 93. Mamane, Y., et al. Interferon regulatory factors: the next generation. Gene 237, 1- 14 (1999). 94. Sato, M., Taniguchi, T. & Tanaka, N. The interferon system and interferon regulatory factor transcription factors -- studies from gene knockout mice. Cytokine Growth Factor Rev 12, 133-142 (2001). 193

95. Taniguchi, T., Ogasawara, K., Takaoka, A. & Tanaka, N. IRF family of transcription factors as regulators of host defense. Annu Rev Immunol 19, 623-655 (2001). 96. de Veer, M.J., et al. Functional classification of interferon-stimulated genes identified using microarrays. J Leukoc Biol 69, 912-920 (2001). 97. Le Page, C., Genin, P., Baines, M.G. & Hiscott, J. Interferon activation and innate immunity. Rev Immunogenet 2, 374-386 (2000). 98. Stark, G.R., Kerr, I.M., Williams, B.R., Silverman, R.H. & Schreiber, R.D. How cells respond to interferons. Annu Rev Biochem 67, 227-264 (1998). 99. Leib, D.A., et al. Interferons regulate the phenotype of wild-type and mutant herpes simplex viruses in vivo. J Exp Med 189, 663-672 (1999). 100. Millan, J. & Ridley, A.J. Rho GTPases and leucocyte-induced endothelial remodelling. Biochem J 385, 329-337 (2005). 101. Davis, J.J. & Fang, B. Oncolytic virotherapy for cancer treatment: challenges and solutions. J Gene Med 7, 1380-1389 (2005). 102. Lamfers, M.L., et al. Cyclophosphamide increases transgene expression mediated by an oncolytic adenovirus in glioma-bearing mice monitored by bioluminescence imaging. Mol Ther 14, 779-788 (2006). 103. Breitbach, C.J., et al. Targeted inflammation during oncolytic virus therapy severely compromises tumor blood flow. Mol Ther 15, 1686-1693 (2007). 104. Orange, J.S. Human natural killer cell deficiencies. Curr Opin Allergy Clin Immunol 6, 399-409 (2006). 105. Orange, J.S. Human natural killer cell deficiencies and susceptibility to infection. Microbes Infect 4, 1545-1558 (2002). 106. Biron, C.A., Byron, K.S. & Sullivan, J.L. Severe herpesvirus infections in an adolescent without natural killer cells. N Engl J Med 320, 1731-1735 (1989). 107. Castriconi, R., et al. NK cells recognize and kill human glioblastoma cells with stem cell-like properties. J Immunol 182, 3530-3539 (2009). 108. Gousias, K., et al. Frequent abnormalities of the immune system in gliomas and correlation with the WHO grading system of malignancy. J Neuroimmunol. 109. Biron, C.A., Nguyen, K.B., Pien, G.C., Cousens, L.P. & Salazar-Mather, T.P. Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu Rev Immunol 17, 189-220 (1999). 110. Yu, J., et al. Pro- and antiinflammatory cytokine signaling: reciprocal antagonism regulates interferon-gamma production by human natural killer cells. Immunity 24, 575-590 (2006). 111. Trotta, R., et al. TGF-beta utilizes SMAD3 to inhibit CD16-mediated IFN-gamma production and antibody-dependent cellular cytotoxicity in human NK cells. J Immunol 181, 3784-3792 (2008). 112. Chisholm, S.E., Howard, K., Gómez, M.V. & Reyburn, H.T. Expression of ICP0 is sufficient to trigger natural killer cell recognition of herpes simplex virus- infected cells by natural cytotoxicity receptors. J Infect Dis 195, 1160-1168 (2007).

194

113. Schepis, D., et al. Herpes simplex virus infection downmodulates NKG2D ligand expression. Scand J Immunol 69, 429-436 (2009). 114. Paulnock, D.M., Demick, K.P. & Coller, S.P. Analysis of interferon-gamma- dependent and -independent pathways of macrophage activation. J Leukoc Biol 67, 677-682 (2000). 115. Wu, A., et al. Expression of MHC I and NK ligands on human CD133+ glioma cells: possible targets of immunotherapy. J Neurooncol 83, 121-131 (2007). 116. Ikeda, K., et al. Complement depletion facilitates the infection of multiple brain tumors by an intravascular, replication-conditional herpes simplex virus mutant. J Virol 74, 4765-4775 (2000). 117. Wakimoto, H., Fulci, G., Tyminski, E. & Chiocca, E.A. Altered expression of antiviral cytokine mRNAs associated with cyclophosphamide's enhancement of viral oncolysis. Gene Ther 11, 214-223 (2004). 118. Ikeda, K., et al. Oncolytic virus therapy of multiple tumors in the brain requires suppression of innate and elicited antiviral responses. Nat Med 5, 881-887 (1999). 119. Otsuki, A., et al. Histone deacetylase inhibitors augment antitumor efficacy of herpes-based oncolytic viruses. Mol Ther 16, 1546-1555 (2008). 120. Katze, M.G., He, Y. & Gale, M., Jr. Viruses and interferon: a fight for supremacy. Nat Rev Immunol 2, 675-687 (2002). 121. Myers, R.M., et al. Preclinical pharmacology and toxicology of intravenous MV- NIS, an oncolytic measles virus administered with or without cyclophosphamide. Clin Pharmacol Ther 82, 700-710 (2007). 122. Louis, D.N., et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 114, 97-109 (2007). 123. Furnari, F.B., et al. Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes Dev 21, 2683-2710 (2007). 124. Mischel, P.S., Nelson, S.F. & Cloughesy, T.F. Molecular analysis of glioblastoma: pathway profiling and its implications for patient therapy. Cancer Biol Ther 2, 242-247 (2003). 125. Phillips, H.S., et al. Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell 9, 157-173 (2006). 126. Maher, E.A., et al. Marked genomic differences characterize primary and secondary glioblastoma subtypes and identify two distinct molecular and clinical secondary glioblastoma entities. Cancer Res 66, 11502-11513 (2006). 127. Lee, J., et al. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines. Cancer Cell 9, 391-403 (2006). 128. Diehn, M., et al. Identification of noninvasive imaging surrogates for brain tumor gene-expression modules. Proc Natl Acad Sci U S A 105, 5213-5218 (2008). 129. Liang, Y., et al. Gene expression profiling reveals molecularly and clinically distinct subtypes of glioblastoma multiforme. Proc Natl Acad Sci U S A 102, 5814-5819 (2005).

195

130. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 455, 1061-1068 (2008). 131. Parsons, D.W., et al. An integrated genomic analysis of human glioblastoma multiforme. Science 321, 1807-1812 (2008). 132. Friese, M.A., et al. MICA/NKG2D-mediated immunogene therapy of experimental gliomas. Cancer Res 63, 8996-9006 (2003). 133. Roth, P., et al. Regeneration and tolerance factor: a novel mediator of glioblastoma-associated immunosuppression. Cancer Res 66, 3852-3858 (2006). 134. Elliott, L.H., Brooks, W.H. & Roszman, T.L. Cytokinetic basis for the impaired activation of lymphocytes from patients with primary intracranial tumors. J Immunol 132, 1208-1215 (1984). 135. Ausiello, C.M., et al. Cell mediated cytotoxicity and cytokine production in peripheral blood mononuclear cells of glioma patients. Eur J Cancer 27, 646-650 (1991). 136. Roussel, E., Gingras, M.C., Grimm, E.A., Bruner, J.M. & Moser, R.P. Predominance of a type 2 intratumoural immune response in fresh tumour- infiltrating lymphocytes from human gliomas. Clin Exp Immunol 105, 344-352 (1996). 137. Zou, J.P., et al. Human glioma-induced immunosuppression involves soluble factor(s) that alters monocyte cytokine profile and surface markers. J Immunol 162, 4882-4892 (1999). 138. Fontana, A., Hengartner, H., de Tribolet, N. & Weber, E. Glioblastoma cells release interleukin 1 and factors inhibiting interleukin 2-mediated effects. J Immunol 132, 1837-1844 (1984). 139. Weller, M., Constam, D.B., Malipiero, U. & Fontana, A. Transforming growth factor-beta 2 induces apoptosis of murine T cell clones without down-regulating bcl-2 mRNA expression. Eur J Immunol 24, 1293-1300 (1994). 140. Hishii, M., et al. Human glioma-derived interleukin-10 inhibits antitumor immune responses in vitro. Neurosurgery 37, 1160-1166; discussion 1166-1167 (1995). 141. Wiendl, H., et al. A functional role of HLA-G expression in human gliomas: an alternative strategy of immune escape. J Immunol 168, 4772-4780 (2002). 142. Wintterle, S., et al. Expression of the B7-related molecule B7-H1 by glioma cells: a potential mechanism of immune paralysis. Cancer Res 63, 7462-7467 (2003). 143. Wu, A., et al. Glioma cancer stem cells induce immunosuppressive macrophages/microglia. Neuro Oncol 12, 1113-1125. 144. Alvarez-Breckenridge, C., Kaur, B. & Chiocca, E.A. Pharmacologic and chemical adjuvants in tumor virotherapy. Chem Rev 109, 3125-3140 (2009). 145. Fulci, G., et al. Depletion of peripheral macrophages and brain microglia increases brain tumor titers of oncolytic viruses. Cancer Res 67, 9398-9406 (2007). 146. Chiocca, E.A. The host response to cancer virotherapy. Curr Opin Mol Ther 10, 38-45 (2008). 147. Hiscott, J. Triggering the innate antiviral response through IRF-3 activation. J Biol Chem 282, 15325-15329 (2007). 196

148. Uematsu, S. & Akira, S. Toll-like receptors and Type I interferons. J Biol Chem 282, 15319-15323 (2007). 149. Mohr, I. Neutralizing innate host defenses to control viral translation in HSV-1 infected cells. Int Rev Immunol 23, 199-220 (2004). 150. Barber, G.N. VSV-tumor selective replication and protein translation. Oncogene 24, 7710-7719 (2005). 151. Hardcastle, J., Kurozumi, K., Chiocca, E.A. & Kaur, B. Oncolytic viruses driven by tumor-specific promoters. Curr Cancer Drug Targets 7, 181-189 (2007). 152. Kim, M., Chung, Y.H. & Johnston, R.N. Reovirus and tumor oncolysis. J Microbiol 45, 187-192 (2007). 153. Johnston, J.B., Nazarian, S.H., Natale, R. & McFadden, G. Myxoma virus infection of primary human fibroblasts varies with cellular age and is regulated by host interferon responses. Virology 332, 235-248 (2005). 154. Kodukula, P., Liu, T., Rooijen, N.V., Jager, M.J. & Hendricks, R.L. Macrophage control of herpes simplex virus type 1 replication in the peripheral nervous system. J Immunol 162, 2895-2905 (1999). 155. Fujioka, N., et al. Interleukin-18 protects mice against acute herpes simplex virus type 1 infection. J Virol 73, 2401-2409 (1999). 156. Burns, A.R., Smith, C.W. & Walker, D.C. Unique structural features that influence neutrophil emigration into the lung. Physiol Rev 83, 309-336 (2003). 157. Pawlik, T.M., et al. Oncolysis of diffuse hepatocellular carcinoma by intravascular administration of a replication-competent, genetically engineered herpesvirus. Cancer Res 60, 2790-2795 (2000). 158. Pawlik, T.M., et al. Prodrug bioactivation and oncolysis of diffuse liver metastases by a herpes simplex virus 1 mutant that expresses the CYP2B1 transgene. Cancer 95, 1171-1181 (2002). 159. Chase, M., Chung, R.Y. & Chiocca, E.A. An oncolytic viral mutant that delivers the CYP2B1 transgene and augments cyclophosphamide chemotherapy. Nat Biotechnol 16, 444-448 (1998). 160. Aghi, M., Chou, T.C., Suling, K., Breakefield, X.O. & Chiocca, E.A. Multimodal cancer treatment mediated by a replicating oncolytic virus that delivers the oxazaphosphorine/rat cytochrome P450 2B1 and ganciclovir/herpes simplex virus thymidine kinase gene therapies. Cancer Res 59, 3861-3865 (1999). 161. Tyminski, E., et al. Brain tumor oncolysis with replication-conditional herpes simplex virus type 1 expressing the prodrug-activating genes, CYP2B1 and secreted human intestinal carboxylesterase, in combination with cyclophosphamide and irinotecan. Cancer Res 65, 6850-6857 (2005). 162. Dingli, D., et al. Image-guided radiovirotherapy for multiple myeloma using a recombinant measles virus expressing the thyroidal sodium iodide symporter. Blood 103, 1641-1646 (2004). 163. Hasegawa, K., et al. Dual therapy of ovarian cancer using measles viruses expressing carcinoembryonic antigen and sodium iodide symporter. Clin Cancer Res 12, 1868-1875 (2006).

197

164. Hellums, E.K., et al. Increased efficacy of an interleukin-12-secreting herpes simplex virus in a syngeneic intracranial murine glioma model. Neuro Oncol 7, 213-224 (2005). 165. Parker, J.N., et al. Engineered herpes simplex virus expressing IL-12 in the treatment of experimental murine brain tumors. Proc Natl Acad Sci U S A 97, 2208-2213 (2000). 166. Ino, Y., Saeki, Y., Fukuhara, H. & Todo, T. Triple combination of oncolytic herpes simplex virus-1 vectors armed with interleukin-12, interleukin-18, or soluble B7-1 results in enhanced antitumor efficacy. Clin Cancer Res 12, 643-652 (2006). 167. Todo, T., Martuza, R.L., Dallman, M.J. & Rabkin, S.D. In situ expression of soluble B7-1 in the context of oncolytic herpes simplex virus induces potent antitumor immunity. Cancer Res 61, 153-161 (2001). 168. Liu, T.C., et al. Dominant-negative fibroblast growth factor receptor expression enhances antitumoral potency of oncolytic herpes simplex virus in neural tumors. Clin Cancer Res 12, 6791-6799 (2006). 169. Liu, T.C., et al. Oncolytic HSV armed with platelet factor 4, an antiangiogenic agent, shows enhanced efficacy. Mol Ther 14, 789-797 (2006). 170. Garber, K. China approves world's first oncolytic virus therapy for cancer treatment. J Natl Cancer Inst 98, 298-300 (2006). 171. Peng, K.W., Facteau, S., Wegman, T., O'Kane, D. & Russell, S.J. Non-invasive in vivo monitoring of trackable viruses expressing soluble marker peptides. Nat Med 8, 527-531 (2002). 172. Peng, K.W., et al. Pharmacokinetics of oncolytic measles virotherapy: eventual equilibrium between virus and tumor in an ovarian cancer xenograft model. Cancer Gene Ther 13, 732-738 (2006). 173. Peng, K.W., et al. Intraperitoneal therapy of ovarian cancer using an engineered measles virus. Cancer Res 62, 4656-4662 (2002). 174. Phuong, L.K., et al. Use of a vaccine strain of measles virus genetically engineered to produce carcinoembryonic antigen as a novel therapeutic agent against glioblastoma multiforme. Cancer Res 63, 2462-2469 (2003). 175. Peng, K.W., et al. Systemic therapy of myeloma xenografts by an attenuated measles virus. Blood 98, 2002-2007 (2001). 176. Steinman, R.M., Turley, S., Mellman, I. & Inaba, K. The induction of tolerance by dendritic cells that have captured apoptotic cells. J Exp Med 191, 411-416 (2000). 177. Gauvrit, A., et al. Measles virus induces oncolysis of mesothelioma cells and allows dendritic cells to cross-prime tumor-specific CD8 response. Cancer Res 68, 4882-4892 (2008). 178. Meng, Y., et al. Successful combination of local CpG-ODN and radiotherapy in malignant glioma. Int J Cancer 116, 992-997 (2005). 179. Li, H., Dutuor, A., Fu, X. & Zhang, X. Induction of strong antitumor immunity by an HSV-2-based oncolytic virus in a murine mammary tumor model. J Gene Med 9, 161-169 (2007). 198

180. Diaz, R.M., et al. Oncolytic immunovirotherapy for melanoma using vesicular stomatitis virus. Cancer Res 67, 2840-2848 (2007). 181. Wakimoto, H., Johnson, P.R., Knipe, D.M. & Chiocca, E.A. Effects of innate immunity on herpes simplex virus and its ability to kill tumor cells. Gene Ther 10, 983-990 (2003). 182. Kambara, H., Saeki, Y. & Chiocca, E.A. Cyclophosphamide allows for in vivo dose reduction of a potent oncolytic virus. Cancer Res 65, 11255-11258 (2005). 183. Li, H., Zeng, Z., Fu, X. & Zhang, X. Coadministration of a herpes simplex virus-2 based oncolytic virus and cyclophosphamide produces a synergistic antitumor effect and enhances tumor-specific immune responses. Cancer Res 67, 7850-7855 (2007). 184. Van Rooijen, N. & Sanders, A. Liposome mediated depletion of macrophages: mechanism of action, preparation of liposomes and applications. J Immunol Methods 174, 83-93 (1994). 185. Nguyen, T.L., et al. Chemical targeting of the innate antiviral response by histone deacetylase inhibitors renders refractory cancers sensitive to viral oncolysis. Proc Natl Acad Sci U S A 105, 14981-14986 (2008). 186. Aghi, M.K. & Chiocca, E.A. Phase ib trial of oncolytic herpes virus G207 shows safety of multiple injections and documents viral replication. Mol Ther 17, 8-9 (2009). 187. Andreansky, S., et al. Treatment of intracranial gliomas in immunocompetent mice using herpes simplex viruses that express murine interleukins. Gene Ther 5, 121-130 (1998). 188. York, I.A., et al. A cytosolic herpes simplex virus protein inhibits antigen presentation to CD8+ T lymphocytes. Cell 77, 525-535 (1994). 189. Hill, A., et al. Herpes simplex virus turns off the TAP to evade host immunity. Nature 375, 411-415 (1995). 190. Fruh, K., et al. A viral inhibitor of peptide transporters for antigen presentation. Nature 375, 415-418 (1995). 191. Todo, T., Martuza, R.L., Rabkin, S.D. & Johnson, P.A. Oncolytic herpes simplex virus vector with enhanced MHC class I presentation and tumor cell killing. Proc Natl Acad Sci U S A 98, 6396-6401 (2001). 192. Bennett, J.J., et al. Interleukin 12 secretion enhances antitumor efficacy of oncolytic herpes simplex viral therapy for colorectal cancer. Ann Surg 233, 819- 826 (2001). 193. Abordo-Adesida, E., et al. Stability of lentiviral vector-mediated transgene expression in the brain in the presence of systemic antivector immune responses. Hum Gene Ther 16, 741-751 (2005). 194. Balachandran, S., Thomas, E. & Barber, G.N. A FADD-dependent innate immune mechanism in mammalian cells. Nature 432, 401-405 (2004). 195. Friedman, A., Tian, J.P., Fulci, G., Chiocca, E.A. & Wang, J. Glioma virotherapy: effects of innate immune suppression and increased viral replication capacity. Cancer Res 66, 2314-2319 (2006).

199

196. Lowenstein, P.R. The case for immunosuppression in clinical gene transfer. Mol Ther 12, 185-186 (2005). 197. Stojdl, D.F., et al. VSV strains with defects in their ability to shutdown innate immunity are potent systemic anti-cancer agents. Cancer Cell 4, 263-275 (2003). 198. Power, A.T., et al. Carrier cell-based delivery of an oncolytic virus circumvents antiviral immunity. Mol Ther 15, 123-130 (2007). 199. Aghi, M., Rabkin, S.D. & Martuza, R.L. Angiogenic response caused by oncolytic herpes simplex virus-induced reduced thrombospondin expression can be prevented by specific viral mutations or by administering a thrombospondin- derived peptide. Cancer Res 67, 440-444 (2007). 200. Ikeda, K., et al. Oncolytic virus therapy of multiple tumors in the brain requires suppression of innate and elicited antiviral responses. Nat Med 5, 881-887 (1999). 201. Wakimoto, H., et al. The complement response against an oncolytic virus is species-specific in its activation pathways. Mol Ther 5, 275-282 (2002). 202. Fulci, G., et al. Cyclophosphamide enhances glioma virotherapy by inhibiting innate immune responses. Proc Natl Acad Sci U S A 103, 12873-12878 (2006). 203. Herrlinger, U., et al. Pre-existing herpes simplex virus 1 (HSV-1) immunity decreases, but does not abolish, gene transfer to experimental brain tumors by a HSV-1 vector. Gene Ther 5, 809-819 (1998). 204. Kurozumi, K., et al. Effect of tumor microenvironment modulation on the efficacy of oncolytic virus therapy. J Natl Cancer Inst 99, 1768-1781 (2007). 205. Rein, D.T., Breidenbach, M. & Curiel, D.T. Current developments in adenovirus- based cancer gene therapy. Future Oncol 2, 137-143 (2006). 206. Hahm, B., et al. Measles virus infects and suppresses proliferation of T lymphocytes from transgenic mice bearing human signaling lymphocytic activation molecule. J Virol 77, 3505-3515 (2003). 207. Tatsuo, H., Ono, N., Tanaka, K. & Yanagi, Y. SLAM (CDw150) is a cellular receptor for measles virus. Nature 406, 893-897 (2000). 208. McQuaid, S. & Cosby, S.L. An immunohistochemical study of the distribution of the measles virus receptors, CD46 and SLAM, in normal human tissues and subacute sclerosing panencephalitis. Lab Invest 82, 403-409 (2002). 209. Shusta, E.V., Zhu, C., Boado, R.J. & Pardridge, W.M. Subtractive expression cloning reveals high expression of CD46 at the blood-brain barrier. J Neuropathol Exp Neurol 61, 597-604 (2002). 210. Allen, C., et al. Interleukin-13 displaying retargeted oncolytic measles virus strains have significant activity against gliomas with improved specificity. Mol Ther 16, 1556-1564 (2008). 211. Caput, D., et al. Cloning and characterization of a specific interleukin (IL)-13 binding protein structurally related to the IL-5 receptor alpha chain. J Biol Chem 271, 16921-16926 (1996). 212. Debinski, W. & Gibo, D.M. Molecular expression analysis of restrictive receptor for interleukin 13, a brain tumor-associated cancer/testis antigen. Mol Med 6, 440- 449 (2000).

200

213. Debinski, W., Gibo, D.M., Hulet, S.W., Connor, J.R. & Gillespie, G.Y. Receptor for interleukin 13 is a marker and therapeutic target for human high-grade gliomas. Clin Cancer Res 5, 985-990 (1999). 214. Menotti, L., et al. Inhibition of human tumor growth in mice by an oncolytic herpes simplex virus designed to target solely HER-2-positive cells. Proc Natl Acad Sci USA 106, 9039-9044 (2009). 215. Grandi, P., et al. Targeting HSV-1 virions for specific binding to epidermal growth factor receptor-vIII-bearing tumor cells. Cancer Gene Ther (2010). 216. Lopez, M., et al. Nectin2alpha (PRR2alpha or HveB) and nectin2delta are low- efficiency mediators for entry of herpes simplex virus mutants carrying the Leu25Pro substitution in glycoprotein D. J Virol 74, 1267-1274 (2000). 217. Akhtar, J. & Shukla, D. Viral entry mechanisms: cellular and viral mediators of herpes simplex virus entry. FEBS J (2009). 218. Krummenacher, C., et al. Comparative usage of herpesvirus entry mediator A and nectin-1 by laboratory strains and clinical isolates of herpes simplex virus. Virology 322, 286-299 (2004). 219. Kawai, T. & Akira, S. Innate immune recognition of viral infection. Nat Immunol 7, 131-137 (2006). 220. Hemmi, H., et al. A Toll-like receptor recognizes bacterial DNA. Nature 408, 740-745 (2000). 221. Perry, A.K., Chow, E.K., Goodnough, J.B., Yeh, W.C. & Cheng, G. Differential requirement for TANK-binding kinase-1 in type I interferon responses to toll-like receptor activation and viral infection. J Exp Med 199, 1651-1658 (2004). 222. Kato, H., et al. Cell type-specific involvement of RIG-I in antiviral response. Immunity 23, 19-28 (2005). 223. Curtin, J.F., et al. Fms-like tyrosine kinase 3 ligand recruits plasmacytoid dendritic cells to the brain. J Immunol 176, 3566-3577 (2006). 224. Plumb, J., Armstrong, M.A., Duddy, M., Mirakhur, M. & McQuaid, S. CD83- positive dendritic cells are present in occasional perivascular cuffs in multiple sclerosis lesions. Mult Scler 9, 142-147 (2003). 225. Serafini, B., Columba-Cabezas, S., Di Rosa, F. & Aloisi, F. Intracerebral recruitment and maturation of dendritic cells in the onset and progression of experimental autoimmune encephalomyelitis. Am J Pathol 157, 1991-2002 (2000). 226. Fischer, H.G. & Reichmann, G. Brain dendritic cells and macrophages/microglia in central nervous system inflammation. J Immunol 166, 2717-2726 (2001). 227. Kumar, K.P., McBride, K.M., Weaver, B.K., Dingwall, C. & Reich, N.C. Regulated nuclear-cytoplasmic localization of interferon regulatory factor 3, a subunit of double-stranded RNA-activated factor 1. Mol Cell Biol 20, 4159-4168 (2000). 228. Lin, R., Genin, P., Mamane, Y. & Hiscott, J. Selective DNA binding and association with the CREB binding protein coactivator contribute to differential activation of alpha/beta interferon genes by interferon regulatory factors 3 and 7. Mol Cell Biol 20, 6342-6353 (2000). 201

229. Sato, M., et al. Positive feedback regulation of type I IFN genes by the IFN- inducible transcription factor IRF-7. FEBS Lett 441, 106-110 (1998). 230. Lin, R., Mamane, Y. & Hiscott, J. Multiple regulatory domains control IRF-7 activity in response to virus infection. J Biol Chem 275, 34320-34327 (2000). 231. Au, W.C., Moore, P.A., LaFleur, D.W., Tombal, B. & Pitha, P.M. Characterization of the interferon regulatory factor-7 and its potential role in the transcription activation of interferon A genes. J Biol Chem 273, 29210-29217 (1998). 232. Marie, I., Durbin, J.E. & Levy, D.E. Differential viral induction of distinct interferon-alpha genes by positive feedback through interferon regulatory factor- 7. EMBO J 17, 6660-6669 (1998). 233. Medzhitov, R. & Janeway, C.A., Jr. Decoding the patterns of self and nonself by the innate immune system. Science 296, 298-300 (2002). 234. Norman, K.L. & Lee, P.W. Not all viruses are bad guys: the case for reovirus in cancer therapy. Drug Discov Today 10, 847-855 (2005). 235. Strong, J.E., Coffey, M.C., Tang, D., Sabinin, P. & Lee, P.W. The molecular basis of viral oncolysis: usurpation of the Ras signaling pathway by reovirus. Embo J 17, 3351-3362 (1998). 236. Strong, J.E. & Lee, P.W. The v-erbB oncogene confers enhanced cellular susceptibility to reovirus infection. J Virol 70, 612-616 (1996). 237. Haralambieva, I., et al. Engineering oncolytic measles virus to circumvent the intracellular innate immune response. Mol Ther 15, 588-597 (2007). 238. McKee, T.D., et al. Degradation of fibrillar collagen in a human melanoma xenograft improves the efficacy of an oncolytic herpes simplex virus vector. Cancer Res 66, 2509-2513 (2006). 239. Shen, B.H., Bauzon, M. & Hermiston, T.W. The effect of hypoxia on the uptake, replication and lytic potential of group B adenovirus type 3 (Ad3) and type 11p (Ad11p). Gene Ther 13, 986-990 (2006). 240. Parkins, C.S., Stratford, M.R., Dennis, M.F., Stubbs, M. & Chaplin, D.J. The relationship between extracellular lactate and tumour pH in a murine tumour model of ischaemia-reperfusion. Br J Cancer 75, 319-323 (1997). 241. Pipiya, T., et al. Hypoxia reduces adenoviral replication in cancer cells by downregulation of viral protein expression. Gene Ther 12, 911-917 (2005). 242. Frantz, S., Vincent, K.A., Feron, O. & Kelly, R.A. Innate immunity and angiogenesis. Circ Res 96, 15-26 (2005). 243. Jackson, J.R., Seed, M.P., Kircher, C.H., Willoughby, D.A. & Winkler, J.D. The codependence of angiogenesis and chronic inflammation. Faseb J 11, 457-465 (1997). 244. Guidotti, L.G. & Chisari, F.V. Noncytolytic control of viral infections by the innate and adaptive immune response. Annu Rev Immunol 19, 65-91 (2001). 245. Baraz, L., Khazanov, E., Condiotti, R., Kotler, M. & Nagler, A. Natural killer (NK) cells prevent virus production in cell culture. Bone Marrow Transplant 24, 179-189 (1999).

202

246. Cooper, M.A., et al. Human natural killer cells: a unique innate immunoregulatory role for the CD56(bright) subset. Blood 97, 3146-3151 (2001). 247. Hickey, W.F. Leukocyte traffic in the central nervous system: the participants and their roles. Semin Immunol 11, 125-137 (1999). 248. Fulci, G. & Chiocca, E.A. Oncolytic viruses for the therapy of brain tumors and other solid malignancies: a review. Front Biosci 8, e346-360 (2003). 249. Markert, J.M., et al. Phase Ib Trial of Mutant Herpes Simplex Virus G207 Inoculated Pre-and Post-tumor Resection for Recurrent GBM. Mol Ther (2008). 250. Hoffmann, D., Bayer, W. & Wildner, O. Local and distant immune-mediated control of colon cancer growth with fusogenic membrane glycoproteins in combination with viral oncolysis. Hum Gene Ther 18, 435-450 (2007). 251. Nakamura, H., et al. Regulation of herpes simplex virus gamma(1)34.5 expression and oncolysis of diffuse liver metastases by Myb34.5. J Clin Invest 109, 871-882 (2002). 252. Grote, D., Cattaneo, R. & Fielding, A.K. Neutrophils contribute to the measles virus-induced antitumor effect: enhancement by granulocyte macrophage colony- stimulating factor expression. Cancer Res 63, 6463-6468 (2003). 253. Prestwich, R., et al. The Case of Oncolytic Viruses vs The Immune System: Waiting On The Judgment of Solomon. Hum Gene Ther (2009). 254. Conrady, C.D., Drevets, D.A. & Carr, D.J. Herpes simplex type I (HSV-1) infection of the nervous system: is an immune response a good thing? J Neuroimmunol 220, 1-9. 255. Shukla, D. & Spear, P.G. Herpesviruses and heparan sulfate: an intimate relationship in aid of viral entry. J Clin Invest 108, 503-510 (2001). 256. Reske, A., Pollara, G., Krummenacher, C., Chain, B.M. & Katz, D.R. Understanding HSV-1 entry glycoproteins. Rev Med Virol 17, 205-215 (2007). 257. Yokota, S., et al. Herpes simplex virus type 1 suppresses the interferon signaling pathway by inhibiting phosphorylation of STATs and janus kinases during an early infection stage. Virology 286, 119-124 (2001). 258. Halford, W.P., et al. ICP0 antagonizes Stat 1-dependent repression of herpes simplex virus: implications for the regulation of viral latency. Virol J 3, 44 (2006). 259. Pasieka, T.J., et al. Herpes simplex virus virion host shutoff attenuates establishment of the antiviral state. J Virol 82, 5527-5535 (2008). 260. Harle, P., Sainz, B., Jr., Carr, D.J. & Halford, W.P. The immediate-early protein, ICP0, is essential for the resistance of herpes simplex virus to interferon- alpha/beta. Virology 293, 295-304 (2002). 261. Lin, R., Noyce, R.S., Collins, S.E., Everett, R.D. & Mossman, K.L. The herpes simplex virus ICP0 RING finger domain inhibits IRF3- and IRF7-mediated activation of interferon-stimulated genes. J Virol 78, 1675-1684 (2004). 262. Melroe, G.T., DeLuca, N.A. & Knipe, D.M. Herpes simplex virus 1 has multiple mechanisms for blocking virus-induced interferon production. J Virol 78, 8411- 8420 (2004).

203

263. Kielian, T. Toll-like receptors in central nervous system glial inflammation and homeostasis. J Neurosci Res 83, 711-730 (2006). 264. Yamamoto, M., et al. Essential role for TIRAP in activation of the signalling cascade shared by TLR2 and TLR4. Nature 420, 324-329 (2002). 265. Akira, S. & Takeda, K. Toll-like receptor signalling. Nat Rev Immunol 4, 499-511 (2004). 266. Akira, S., Uematsu, S. & Takeuchi, O. Pathogen recognition and innate immunity. Cell 124, 783-801 (2006). 267. O'Neill, L.A. & Bowie, A.G. The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nat Rev Immunol 7, 353-364 (2007). 268. Ahmad, R., El Bassam, S., Cordeiro, P. & Menezes, J. Requirement of TLR2- mediated signaling for the induction of IL-15 gene expression in human monocytic cells by HSV-1. Blood 112, 2360-2368 (2008). 269. Sato, A., Linehan, M.M. & Iwasaki, A. Dual recognition of herpes simplex viruses by TLR2 and TLR9 in dendritic cells. Proc Natl Acad Sci U S A 103, 17343-17348 (2006). 270. Sorensen, L.N., et al. TLR2 and TLR9 synergistically control herpes simplex virus infection in the brain. J Immunol 181, 8604-8612 (2008). 271. Krug, A., et al. Herpes simplex virus type 1 activates murine natural interferon- producing cells through toll-like receptor 9. Blood 103, 1433-1437 (2004). 272. Kawai, T., et al. Interferon-alpha induction through Toll-like receptors involves a direct interaction of IRF7 with MyD88 and TRAF6. Nat Immunol 5, 1061-1068 (2004). 273. Yang, K., et al. Human TLR-7-, -8-, and -9-mediated induction of IFN-alpha/beta and -lambda Is IRAK-4 dependent and redundant for protective immunity to viruses. Immunity 23, 465-478 (2005). 274. Rasmussen, S.B., et al. Type I interferon production during herpes simplex virus infection is controlled by cell-type-specific viral recognition through Toll-like receptor 9, the mitochondrial antiviral signaling protein pathway, and novel recognition systems. J Virol 81, 13315-13324 (2007). 275. Alexopoulou, L., Holt, A.C., Medzhitov, R. & Flavell, R.A. Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 413, 732-738 (2001). 276. Doyle, S., et al. IRF3 mediates a TLR3/TLR4-specific antiviral gene program. Immunity 17, 251-263 (2002). 277. Boehme, K.W. & Compton, T. Innate sensing of viruses by toll-like receptors. J Virol 78, 7867-7873 (2004). 278. Onoguchi, K., et al. Viral infections activate types I and III interferon genes through a common mechanism. J Biol Chem 282, 7576-7581 (2007). 279. Darnell, J.E., Jr., Kerr, I.M. & Stark, G.R. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264, 1415-1421 (1994).

204

280. Zhou, A., Hassel, B.A. & Silverman, R.H. Expression cloning of 2-5A-dependent RNAase: a uniquely regulated mediator of interferon action. Cell 72, 753-765 (1993). 281. Samuel, C.E. Antiviral actions of interferons. Clin Microbiol Rev 14, 778-809, table of contents (2001). 282. Mossman, K.L. & Ashkar, A.A. Herpesviruses and the innate immune response. Viral Immunol 18, 267-281 (2005). 283. Larder, B.A., Kemp, S.D. & Darby, G. Related functional domains in virus DNA polymerases. EMBO J 6, 169-175 (1987). 284. Moore, K.W., et al. Homology of cytokine synthesis inhibitory factor (IL-10) to the Epstein-Barr virus gene BCRFI. Science 248, 1230-1234 (1990). 285. Chou, J. & Roizman, B. Herpes simplex virus 1 gamma(1)34.5 gene function, which blocks the host response to infection, maps in the homologous domain of the genes expressed during growth arrest and DNA damage. Proc Natl Acad Sci U S A 91, 5247-5251 (1994). 286. Selvakumaran, M., Reed, J.C., Liebermann, D. & Hoffman, B. Progression of the myeloid differentiation program is dominant to transforming growth factor-beta 1-induced apoptosis in M1 myeloid leukemic cells. Blood 84, 1036-1042 (1994). 287. Zhan, Q., et al. The gadd and MyD genes define a novel set of mammalian genes encoding acidic proteins that synergistically suppress cell growth. Mol Cell Biol 14, 2361-2371 (1994). 288. He, B., Chou, J., Liebermann, D.A., Hoffman, B. & Roizman, B. The carboxyl terminus of the murine MyD116 gene substitutes for the corresponding domain of the gamma(1)34.5 gene of herpes simplex virus to preclude the premature shutoff of total protein synthesis in infected human cells. J Virol 70, 84-90 (1996). 289. Veerapong, J., et al. Systemic delivery of (gamma1)34.5-deleted herpes simplex virus-1 selectively targets and treats distant human xenograft tumors that express high MEK activity. Cancer Res 67, 8301-8306 (2007). 290. Favoreel, H.W., Van de Walle, G.R., Nauwynck, H.J. & Pensaert, M.B. Virus complement evasion strategies. J Gen Virol 84, 1-15 (2003). 291. Ghiringhelli, F., et al. CD4+CD25+ regulatory T cells suppress tumor immunity but are sensitive to cyclophosphamide which allows immunotherapy of established tumors to be curative. Eur J Immunol 34, 336-344 (2004). 292. Ghiringhelli, F., et al. Metronomic cyclophosphamide regimen selectively depletes CD4+CD25+ regulatory T cells and restores T and NK effector functions in end stage cancer patients. Cancer Immunol Immunother 56, 641-648 (2007). 293. Di Paolo, N.C., et al. Effect of adenovirus-mediated heat shock protein expression and oncolysis in combination with low-dose cyclophosphamide treatment on antitumor immune responses. Cancer Res 66, 960-969 (2006). 294. Glaser, M. Augmentation of specific immune response against a syngeneic SV40- induced sarcoma in mice by depletion of suppressor T cells with cyclophosphamide. Cell Immunol 48, 339-345 (1979).

205

295. Rollinghoff, M., Starzinski-Powitz, A., Pfizenmaier, K. & Wagner, H. Cyclophosphamide-sensitive T lymphocytes suppress the in vivo generation of antigen-specific cytotoxic T lymphocytes. J Exp Med 145, 455-459 (1977). 296. Beyer, M., et al. Reduced frequencies and suppressive function of CD4+CD25hi regulatory T cells in patients with chronic lymphocytic leukemia after therapy with fludarabine. Blood 106, 2018-2025 (2005). 297. Ikezawa, Y., et al. Cyclophosphamide decreases the number, percentage and the function of CD25+ CD4+ regulatory T cells, which suppress induction of contact hypersensitivity. J Dermatol Sci 39, 105-112 (2005). 298. Lutsiak, M.E., et al. Inhibition of CD4(+)25+ T regulatory cell function implicated in enhanced immune response by low-dose cyclophosphamide. Blood 105, 2862-2868 (2005). 299. Kottke, T., et al. Improved systemic delivery of oncolytic reovirus to established tumors using preconditioning with cyclophosphamide-mediated Treg modulation and interleukin-2. Clin Cancer Res 15, 561-569 (2009). 300. Qiao, J., et al. Cyclophosphamide facilitates antitumor efficacy against subcutaneous tumors following intravenous delivery of reovirus. Clin Cancer Res 14, 259-269 (2008). 301. Chang, H.M., et al. Induction of interferon-stimulated gene expression and antiviral responses require protein deacetylase activity. Proc Natl Acad Sci U S A 101, 9578-9583 (2004). 302. Nusinzon, I. & Horvath, C.M. Interferon-stimulated transcription and innate antiviral immunity require deacetylase activity and histone deacetylase 1. Proc Natl Acad Sci U S A 100, 14742-14747 (2003). 303. Vlasakova, J., et al. Histone deacetylase inhibitors suppress IFNalpha-induced up- regulation of promyelocytic leukemia protein. Blood 109, 1373-1380 (2007). 304. Marks, P.A., Richon, V.M. & Rifkind, R.A. Histone deacetylase inhibitors: inducers of differentiation or apoptosis of transformed cells. J Natl Cancer Inst 92, 1210-1216 (2000). 305. He, L.Z., et al. Histone deacetylase inhibitors induce remission in transgenic models of therapy-resistant acute promyelocytic leukemia. J Clin Invest 108, 1321-1330 (2001). 306. Johnstone, R.W. Histone-deacetylase inhibitors: novel drugs for the treatment of cancer. Nat Rev Drug Discov 1, 287-299 (2002). 307. Kim, M.S., et al. Histone deacetylases induce angiogenesis by negative regulation of tumor suppressor genes. Nat Med 7, 437-443 (2001). 308. Fouladi, M. Histone deacetylase inhibitors in cancer therapy. Cancer Invest 24, 521-527 (2006). 309. Marks, P., et al. Histone deacetylases and cancer: causes and therapies. Nat Rev Cancer 1, 194-202 (2001). 310. Dowdy, S.C., et al. Histone deacetylase inhibitors and paclitaxel cause synergistic effects on apoptosis and microtubule stabilization in papillary serous endometrial cancer cells. Mol Cancer Ther 5, 2767-2776 (2006).

206

311. Qi, H. & Ratnam, M. Synergistic induction of folate receptor beta by all-trans retinoic acid and histone deacetylase inhibitors in acute myelogenous leukemia cells: mechanism and utility in enhancing selective growth inhibition by antifolates. Cancer Res 66, 5875-5882 (2006). 312. Piacentini, P., et al. Trichostatin A enhances the response of chemotherapeutic agents in inhibiting pancreatic cancer cell proliferation. Virchows Arch 448, 797- 804 (2006). 313. Genin, P., Morin, P. & Civas, A. Impairment of interferon-induced IRF-7 gene expression due to inhibition of ISGF3 formation by trichostatin A. J Virol 77, 7113-7119 (2003). 314. Joseph, J., et al. Expression profiling of sodium butyrate (NaB)-treated cells: identification of regulation of genes related to cytokine signaling and cancer metastasis by NaB. Oncogene 23, 6304-6315 (2004). 315. Kelly, W.K. & Marks, P.A. Drug insight: Histone deacetylase inhibitors-- development of the new targeted anticancer agent suberoylanilide hydroxamic acid. Nat Clin Pract Oncol 2, 150-157 (2005). 316. Mehnert, J.M. & Kelly, W.K. Histone deacetylase inhibitors: biology and mechanism of action. Cancer J 13, 23-29 (2007). 317. Minucci, S. & Pelicci, P.G. Histone deacetylase inhibitors and the promise of epigenetic (and more) treatments for cancer. Nat Rev Cancer 6, 38-51 (2006). 318. Nusinzon, I. & Horvath, C.M. Positive and negative regulation of the innate antiviral response and beta interferon gene expression by deacetylation. Mol Cell Biol 26, 3106-3113 (2006). 319. Taplin, M.E. Drug insight: role of the androgen receptor in the development and progression of prostate cancer. Nat Clin Pract Oncol 4, 236-244 (2007). 320. Knudsen, K.E., Diehl, J.A., Haiman, C.A. & Knudsen, E.S. Cyclin D1: polymorphism, aberrant splicing and cancer risk. Oncogene 25, 1620-1628 (2006). 321. Alao, J.P. The regulation of cyclin D1 degradation: roles in cancer development and the potential for therapeutic invention. Mol Cancer 6, 24 (2007). 322. Hu, J. & Colburn, N.H. Histone deacetylase inhibition down-regulates cyclin D1 transcription by inhibiting nuclear factor-kappaB/p65 DNA binding. Mol Cancer Res 3, 100-109 (2005). 323. Liu, T.C., Castelo-Branco, P., Rabkin, S.D. & Martuza, R.L. Trichostatin A and oncolytic HSV combination therapy shows enhanced antitumoral and antiangiogenic effects. Mol Ther 16, 1041-1047 (2008). 324. Kitazono, M., Goldsmith, M.E., Aikou, T., Bates, S. & Fojo, T. Enhanced adenovirus transgene expression in malignant cells treated with the histone deacetylase inhibitor FR901228. Cancer Res 61, 6328-6330 (2001). 325. Okada, T., et al. A histone deacetylase inhibitor enhances recombinant adeno- associated virus-mediated gene expression in tumor cells. Mol Ther 13, 738-746 (2006).

207

326. Bieler, A., et al. Novel three-pronged strategy to enhance cancer cell killing in glioblastoma cell lines: histone deacetylase inhibitor, chemotherapy, and oncolytic adenovirus dl520. Hum Gene Ther 17, 55-70 (2006). 327. Kurozumi, K., et al. Oncolytic HSV-1 Infection of Tumors Induces Angiogenesis and Upregulates CYR61. Mol Ther (2008). 328. Aghi, M.K., Liu, T.C., Rabkin, S. & Martuza, R.L. Hypoxia enhances the replication of oncolytic herpes simplex virus. Mol Ther 17, 51-56 (2009). 329. Arap, W., Pasqualini, R. & Ruoslahti, E. Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model. Science 279, 377-380 (1998). 330. Brooks, P.C., Clark, R.A. & Cheresh, D.A. Requirement of vascular integrin alpha v beta 3 for angiogenesis. Science 264, 569-571 (1994). 331. Rhim, J.H. & Tosato, G. Targeting the tumor vasculature to improve the efficacy of oncolytic virus therapy. J Natl Cancer Inst 99, 1739-1741 (2007). 332. Yamada, S., et al. Effect of the angiogenesis inhibitor Cilengitide (EMD 121974) on glioblastoma growth in nude mice. Neurosurgery 59, 1304-1312; discussion 1312 (2006). 333. Wahl, S.M., et al. Synthetic fibronectin peptides suppress arthritis in rats by interrupting leukocyte adhesion and recruitment. J Clin Invest 94, 655-662 (1994). 334. Elitok, S., et al. Cyclic arginine-glycine-aspartic acid peptide inhibits macrophage infiltration of the kidney and carotid artery lesions in apo-E-deficient mice. Am J Physiol Renal Physiol 290, F159-166 (2006). 335. Zhang, X., et al. Atomic force microscopy measurement of leukocyte-endothelial interaction. Am J Physiol Heart Circ Physiol 286, H359-367 (2004). 336. Kim, S., et al. In vivo developmental stages in murine natural killer cell maturation. Nat Immunol 3, 523-528 (2002). 337. Anfossi, N., et al. Human NK cell education by inhibitory receptors for MHC class I. Immunity 25, 331-342 (2006). 338. Cooper, M.A., Fehniger, T.A. & Caligiuri, M.A. The biology of human natural killer-cell subsets. Trends Immunol 22, 633-640 (2001). 339. Ferlazzo, G. & Munz, C. NK cell compartments and their activation by dendritic cells. J Immunol 172, 1333-1339 (2004). 340. Hayakawa, Y. & Smyth, M.J. CD27 dissects mature NK cells into two subsets with distinct responsiveness and migratory capacity. J Immunol 176, 1517-1524 (2006). 341. Vivier, E., Tomasello, E., Baratin, M., Walzer, T. & Ugolini, S. Functions of natural killer cells. Nat Immunol 9, 503-510 (2008). 342. Walzer, T., Dalod, M., Robbins, S.H., Zitvogel, L. & Vivier, E. Natural-killer cells and dendritic cells: "l'union fait la force". Blood 106, 2252-2258 (2005). 343. Lucas, M., Schachterle, W., Oberle, K., Aichele, P. & Diefenbach, A. Dendritic cells prime natural killer cells by trans-presenting interleukin 15. Immunity 26, 503-517 (2007).

208

344. de Vries, E., et al. Identification of an unusual Fc gamma receptor IIIa (CD16) on natural killer cells in a patient with recurrent infections. Blood 88, 3022-3027 (1996). 345. Jawahar, S., et al. Natural Killer (NK) cell deficiency associated with an epitope- deficient Fc receptor type IIIA (CD16-II). Clin Exp Immunol 103, 408-413 (1996). 346. Lopez, C., et al. Correlation between low natural killing of fibroblasts infected with herpes simplex virus type 1 and susceptibility to herpesvirus infections. J Infect Dis 147, 1030-1035 (1983). 347. Klein, C., et al. Partial albinism with immunodeficiency (Griscelli syndrome). J Pediatr 125, 886-895 (1994). 348. Mancini, A.J., Chan, L.S. & Paller, A.S. Partial albinism with immunodeficiency: Griscelli syndrome: report of a case and review of the literature. J Am Acad Dermatol 38, 295-300 (1998). 349. Plebani, A., Ciravegna, B., Ponte, M., Mingari, M.C. & Moretta, L. Interleukin-2 mediated restoration of natural killer cell function in a patient with Griscelli syndrome. Eur J Pediatr 159, 713-714 (2000). 350. Anderson, D.C., et al. The severe and moderate phenotypes of heritable Mac-1, LFA-1 deficiency: their quantitative definition and relation to leukocyte dysfunction and clinical features. J Infect Dis 152, 668-689 (1985). 351. Lau, Y.L., Low, L.C., Jones, B.M. & Lawton, J.W. Defective neutrophil and lymphocyte function in leucocyte adhesion deficiency. Clin Exp Immunol 85, 202-208 (1991). 352. Kishimoto, T.K., Hollander, N., Roberts, T.M., Anderson, D.C. & Springer, T.A. Heterogeneous mutations in the beta subunit common to the LFA-1, Mac-1, and p150,95 glycoproteins cause leukocyte adhesion deficiency. Cell 50, 193-202 (1987). 353. Shibuya, K., et al. Physical and functional association of LFA-1 with DNAM-1 adhesion molecule. Immunity 11, 615-623 (1999). 354. Wen, P.Y., et al. Medical management of patients with brain tumors. J Neurooncol 80, 313-332 (2006). 355. Chi, A.S. & Wen, P.Y. Inhibiting kinases in malignant gliomas. Expert Opin Ther Targets 11, 473-496 (2007). 356. Sathornsumetee, S., et al. Molecularly targeted therapy for malignant glioma. Cancer 110, 13-24 (2007). 357. Haseley, A., Alvarez-Breckenridge, C., Chaudhury, A.R. & Kaur, B. Advances in oncolytic virus therapy for glioma. Recent patents on CNS drug discovery 4, 1-13 (2009). 358. Agarwalla, P.K., Barnard, Z.R. & Curry, W.T., Jr. Virally mediated immunotherapy for brain tumors. Neurosurg Clin N Am 21, 167-179. 359. Benencia, F., et al. HSV oncolytic therapy upregulates interferon-inducible chemokines and recruits immune effector cells in ovarian cancer. Mol Ther 12, 789-802 (2005).

209

360. Altomonte, J., et al. Enhanced oncolytic potency of vesicular stomatitis virus through vector-mediated inhibition of NK and NKT cells. Cancer Gene Ther (2008). 361. Errington, F., et al. Reovirus Activates Human Dendritic Cells to Promote Innate Antitumor Immunity. J Immunol 180, 6018-6026 (2008). 362. Errington, F., et al. Inflammatory tumour cell killing by oncolytic reovirus for the treatment of melanoma. Gene Ther 15, 1257-1270 (2008). 363. Kottke, T., et al. Use of Biological Therapy to Enhance Both Virotherapy and Adoptive T-Cell Therapy for Cancer. Mol Ther (2008). 364. Prestwich, R., et al. Reciprocal Human Dendritic Cell-Natural Killer Cell Interactions Induce Antitumor Activity Following Tumor Cell Infection by Oncolytic Reovirus. J Immunol (2009). 365. Ali, S., et al. Combined immunostimulation and conditional cytotoxic gene therapy provide long-term survival in a large glioma model. Cancer Res 65, 7194- 7204 (2005). 366. Varghese, S., Rabkin, S.D., Nielsen, P.G., Wang, W. & Martuza, R.L. Systemic oncolytic herpes virus therapy of poorly immunogenic prostate cancer metastatic to lung. Clin Cancer Res 12, 2919-2927 (2006). 367. Farrell, C.J., et al. Combination immunotherapy for tumors via sequential intratumoral injections of oncolytic herpes simplex virus 1 and immature dendritic cells. Clin Cancer Res 14, 7711-7716 (2008). 368. Derubertis, B.G., et al. Cytokine-secreting herpes viral mutants effectively treat tumor in a murine metastatic colorectal liver model by oncolytic and T-cell- dependent mechanisms. Cancer Gene Ther 14, 590-597 (2007). 369. Prestwich, R.J., et al. The case of oncolytic viruses versus the immune system: waiting on the judgment of Solomon. Hum Gene Ther 20, 1119-1132 (2009). 370. Parato, K.A., Lichty, B.D. & Bell, J.C. Diplomatic immunity: turning a foe into an ally. Curr Opin Mol Ther 11, 13-21 (2009). 371. Stanford, M.M., Breitbach, C.J., Bell, J.C. & McFadden, G. Innate immunity, tumor microenvironment and oncolytic virus therapy: friends or foes? Curr Opin Mol Ther 10, 32-37 (2008). 372. Lowenstein, P.R., Mandel, R.J., Xiong, W.D., Kroeger, K. & Castro, M.G. Immune responses to adenovirus and adeno-associated vectors used for gene therapy of brain diseases: the role of immunological synapses in understanding the cell biology of neuroimmune interactions. Curr Gene Ther 7, 347-360 (2007). 373. Chisholm, S.E. & Reyburn, H.T. Recognition of vaccinia virus-infected cells by human natural killer cells depends on natural cytotoxicity receptors. J Virol 80, 2225-2233 (2006). 374. Reilly, K.M., Loisel, D.A., Bronson, R.T., McLaughlin, M.E. & Jacks, T. Nf1;Trp53 mutant mice develop glioblastoma with evidence of strain-specific effects. Nat Genet 26, 109-113 (2000). 375. Marques, C.P., et al. Prolonged microglial cell activation and lymphocyte infiltration following experimental herpes encephalitis. J Immunol 181, 6417- 6426 (2008). 210

376. Kambara, H., Okano, H., Chiocca, E.A. & Saeki, Y. An oncolytic HSV-1 mutant expressing ICP34.5 under control of a nestin promoter increases survival of animals even when symptomatic from a brain tumor. Cancer Res 65, 2832-2839 (2005). 377. Moretta, A., et al. CD69-mediated pathway of lymphocyte activation: anti-CD69 monoclonal antibodies trigger the cytolytic activity of different lymphoid effector cells with the exception of cytolytic T lymphocytes expressing T cell receptor alpha/beta. J Exp Med 174, 1393-1398 (1991). 378. Chen, S., Kawashima, H., Lowe, J.B., Lanier, L.L. & Fukuda, M. Suppression of tumor formation in lymph nodes by L-selectin-mediated natural killer cell recruitment. J Exp Med 202, 1679-1689 (2005). 379. Wu, J., et al. An activating immunoreceptor complex formed by NKG2D and DAP10. Science 285, 730-732 (1999). 380. Smith, K.M., Wu, J., Bakker, A.B., Phillips, J.H. & Lanier, L.L. Ly-49D and Ly- 49H associate with mouse DAP12 and form activating receptors. J Immunol 161, 7-10 (1998). 381. Chiossone, L., et al. Maturation of mouse NK cells is a 4-stage developmental program. Blood 113, 5488-5496 (2009). 382. van Dommelen, S.L., et al. Perforin and granzymes have distinct roles in defensive immunity and immunopathology. Immunity 25, 835-848 (2006). 383. Wang, J.M., Griffin, J.D., Rambaldi, A., Chen, Z.G. & Mantovani, A. Induction of monocyte migration by recombinant macrophage colony-stimulating factor. J Immunol 141, 575-579 (1988). 384. Byrne, P., McGuirk, P., Todryk, S. & Mills, K.H. Depletion of NK cells results in disseminating lethal infection with Bordetella pertussis associated with a reduction of antigen-specific Th1 and enhancement of Th2, but not Tr1 cells. Eur J Immunol 34, 2579-2588 (2004). 385. Sayeh, E., Sterling, K., Speck, E., Freedman, J. & Semple, J.W. IgG antiplatelet immunity is dependent on an early innate natural killer cell-derived interferon- gamma response that is regulated by CD8+ T cells. Blood 103, 2705-2709 (2004). 386. Yoshino, H., et al. Natural killer cell depletion by anti-asialo GM1 antiserum treatment enhances human hematopoietic stem cell engraftment in NOD/Shi-scid mice. Bone Marrow Transplant 26, 1211-1216 (2000). 387. Savarin, C. & Bergmann, C.C. Neuroimmunology of central nervous system viral infections: the cells, molecules and mechanisms involved. Curr Opin Pharmacol 8, 472-479 (2008). 388. Carpentier, A.F., Xie, J., Mokhtari, K. & Delattre, J.Y. Successful treatment of intracranial gliomas in rat by oligodeoxynucleotides containing CpG motifs. Clin Cancer Res 6, 2469-2473 (2000). 389. Altomonte, J., et al. Exponential Enhancement of Oncolytic Vesicular Stomatitis Virus Potency by Vector-mediated Suppression of Inflammatory Responses In Vivo. Mol Ther 16, 146-153 (2008). 390. Kambara, H., Saeki, Y. & Chiocca, E.A. Cyclophosphamide Allows for In vivo Dose Reduction of a Potent Oncolytic Virus. Cancer Research (2005). 211

391. Galivo, F., et al. Interference of CD40L-mediated tumor immunotherapy by oncolytic vesicular stomatitis virus. Hum Gene Ther 21, 439-450. 392. Galivo, F., et al. Single-cycle viral gene expression, rather than progressive replication and oncolysis, is required for VSV therapy of B16 melanoma. Gene Ther 17, 158-170. 393. Sørensen, L.N., et al. TLR2 and TLR9 synergistically control herpes simplex virus infection in the brain. J Immunol 181, 8604-8612 (2008). 394. Chesler, D.A. & Reiss, C.S. The role of IFN-gamma in immune responses to viral infections of the central nervous system. Cytokine Growth Factor Rev 13, 441- 454 (2002). 395. Marques, C.P., Hu, S., Sheng, W. & Lokensgard, J.R. Microglial cells initiate vigorous yet non-protective immune responses during HSV-1 brain infection. Virus Res 121, 1-10 (2006). 396. Lundberg, P., Openshaw, H., Wang, M., Yang, H.J. & Cantin, E. Effects of CXCR3 signaling on development of fatal encephalitis and corneal and periocular skin disease in HSV-infected mice are mouse-strain dependent. Invest Ophthalmol Vis Sci 48, 4162-4170 (2007). 397. Yu, J., et al. CD94 surface density identifies a functional intermediary between the CD56bright and CD56dim human NK-cell subsets. Blood 115, 274-281. 398. Lundberg, P., et al. A locus on mouse chromosome 6 that determines resistance to herpes simplex virus also influences reactivation, while an unlinked locus augments resistance of female mice. Journal of virology 77, 11661-11673 (2003). 399. Kastrukoff, L.F., et al. Redundancy in the immune system restricts the spread of HSV-1 in the central nervous system (CNS) of C57BL/6 mice. Virology 400, 248- 258 (2010). 400. Castriconi, R., et al. Both CD133+ and CD133- medulloblastoma cell lines express ligands for triggering NK receptors and are susceptible to NK-mediated cytotoxicity. Eur J Immunol 37, 3190-3196 (2007). 401. Giannini, C., et al. Patient tumor EGFR and PDGFRA gene amplifications retained in an invasive intracranial xenograft model of glioblastoma multiforme. Neuro Oncol 7, 164-176 (2005). 402. Fuller, C.L., et al. NKp30-dependent cytolysis of filovirus-infected human dendritic cells. Cell Microbiol 9, 962-976 (2007). 403. Shafer-Weaver, K., et al. The Granzyme B ELISPOT assay: an alternative to the 51Cr-release assay for monitoring cell-mediated cytotoxicity. Journal of translational medicine 1, 14 (2003). 404. Vitale, M., et al. NK-dependent DC maturation is mediated by TNFalpha and IFNgamma released upon engagement of the NKp30 triggering receptor. Blood 106, 566-571 (2005). 405. Walzer, T., et al. Identification, activation, and selective in vivo ablation of mouse NK cells via NKp46. Proc Natl Acad Sci U S A 104, 3384-3389 (2007). 406. Becknell, B. & Caligiuri, M.A. Interleukin-2, interleukin-15, and their roles in human natural killer cells. Adv Immunol 86, 209-239 (2005).

212

407. Benson, D.M., Jr., et al. Stem cell factor and interleukin-2/15 combine to enhance MAPK-mediated proliferation of human natural killer cells. Blood 113, 2706- 2714 (2009). 408. Carson, W.E., et al. A potential role for interleukin-15 in the regulation of human natural killer cell survival. J Clin Invest 99, 937-943 (1997). 409. Carson, W.E., et al. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J Exp Med 180, 1395- 1403 (1994). 410. Fehniger, T.A., et al. Cutting edge: IL-15 costimulates the generalized Shwartzman reaction and innate immune IFN-gamma production in vivo. J Immunol 164, 1643-1647 (2000). 411. Mrozek, E., Anderson, P. & Caligiuri, M.A. Role of interleukin-15 in the development of human CD56+ natural killer cells from CD34+ hematopoietic progenitor cells. Blood 87, 2632-2640 (1996). 412. Lewinsohn, D.M., et al. Human purified protein derivative-specific CD4+ T cells use both CD95-dependent and CD95-independent cytolytic mechanisms. J Immunol 160, 2374-2379 (1998). 413. Taylor, M.A., Ward, B., Schatzle, J.D. & Bennett, M. Perforin- and Fas- dependent mechanisms of natural killer cell-mediated rejection of incompatible bone marrow cell grafts. Eur J Immunol 32, 793-799 (2002). 414. Trotta, R., et al. Dependence of both spontaneous and antibody-dependent, granule exocytosis-mediated NK cell cytotoxicity on extracellular signal- regulated kinases. J Immunol 161, 6648-6656 (1998). 415. Esser, M.T., et al. IL-2 induces Fas ligand/Fas (CD95L/CD95) cytotoxicity in CD8+ and CD4+ T lymphocyte clones. J Immunol 158, 5612-5618 (1997). 416. Esser, M.T., Haverstick, D.M., Fuller, C.L., Gullo, C.A. & Braciale, V.L. Ca2+ signaling modulates cytolytic T lymphocyte effector functions. J Exp Med 187, 1057-1067 (1998). 417. Fuller, C.L. & Braciale, V.L. Selective induction of CD8+ cytotoxic T lymphocyte effector function by staphylococcus enterotoxin B. J Immunol 161, 5179-5186 (1998). 418. Brandt, C.S., et al. The B7 family member B7-H6 is a tumor cell ligand for the activating natural killer cell receptor NKp30 in humans. J Exp Med (2009). 419. Vitale, M., et al. NKp44, a novel triggering surface molecule specifically expressed by activated natural killer cells, is involved in non-major histocompatibility complex-restricted tumor cell lysis. J Exp Med 187, 2065-2072 (1998). 420. Moretta, A., et al. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu Rev Immunol 19, 197-223 (2001). 421. Arnon, T.I., Markel, G. & Mandelboim, O. Tumor and viral recognition by natural killer cells receptors. Semin Cancer Biol 16, 348-358 (2006). 422. Bloushtain, N., et al. Membrane-associated heparan sulfate proteoglycans are involved in the recognition of cellular targets by NKp30 and NKp46. J Immunol 173, 2392-2401 (2004). 213

423. Ferlazzo, G., et al. Human dendritic cells activate resting natural killer (NK) cells and are recognized via the NKp30 receptor by activated NK cells. J Exp Med 195, 343-351 (2002). 424. Mailliard, R.B., et al. Dendritic cells mediate NK cell help for Th1 and CTL responses: two-signal requirement for the induction of NK cell helper function. J Immunol 171, 2366-2373 (2003). 425. Arnon, T.I., et al. Inhibition of the NKp30 activating receptor by pp65 of human cytomegalovirus. Nat Immunol 6, 515-523 (2005). 426. Degli-Esposti, M.A. & Smyth, M.J. Close encounters of different kinds: dendritic cells and NK cells take centre stage. Nat Rev Immunol 5, 112-124 (2005). 427. Pogge von Strandmann, E., et al. Human leukocyte antigen-B-associated transcript 3 is released from tumor cells and engages the NKp30 receptor on natural killer cells. Immunity 27, 965-974 (2007). 428. Aghi, M., Rabkin, S. & Martuza, R.L. Effect of chemotherapy-induced DNA repair on oncolytic herpes simplex viral replication. J Natl Cancer Inst 98, 38-50 (2006). 429. Augugliaro, R., et al. Selective cross-talk among natural cytotoxicity receptors in human natural killer cells. Eur J Immunol 33, 1235-1241 (2003). 430. Thorne, S.H., et al. Targeting localized immune suppression within the tumor through repeat cycles of immune cell-oncolytic virus combination therapy. Mol Ther 18, 1698-1705. 431. Ahmed, A.U., et al. Bone marrow mesenchymal stem cells loaded with an oncolytic adenovirus suppress the anti-adenoviral immune response in the cotton rat model. Mol Ther 18, 1846-1856. 432. Lun, X., et al. Myxoma virus virotherapy for glioma in immunocompetent animal models: optimizing administration routes and synergy with rapamycin. Cancer Res 70, 598-608. 433. Kircher, B., et al. Anti-leukemic activity of valproic acid and imatinib mesylate on human Ph+ ALL and CML cells in vitro. Eur J Haematol 83, 48-56 (2009). 434. Chen, W.Y., Bailey, E.C., McCune, S.L., Dong, J.Y. & Townes, T.M. Reactivation of silenced, virally transduced genes by inhibitors of histone deacetylase. Proc Natl Acad Sci U S A 94, 5798-5803 (1997). 435. Dion, L.D., et al. Amplification of recombinant adenoviral transgene products occurs by inhibition of histone deacetylase. Virology 231, 201-209 (1997). 436. Yamano, T., et al. Amplification of transgene expression in vitro and in vivo using a novel inhibitor of histone deacetylase. Mol Ther 1, 574-580 (2000). 437. Currier, M.A., et al. Efficacy and safety of the oncolytic herpes simplex virus rRp450 alone and combined with cyclophosphamide. Mol Ther 16, 879-885 (2008). 438. Nguyên, T.L., et al. Chemical targeting of the innate antiviral response by histone deacetylase inhibitors renders refractory cancers sensitive to viral oncolysis. Proc Natl Acad Sci USA (2008).

214

439. Walzer, T., Dalod, M., Vivier, E. & Zitvogel, L. Natural killer cell-dendritic cell crosstalk in the initiation of immune responses. Expert opinion on biological therapy 5 Suppl 1, S49-59 (2005). 440. Zhang, C., Wang, Y., Zhou, Z., Zhang, J. & Tian, Z. Sodium butyrate upregulates expression of NKG2D ligand MICA/B in HeLa and HepG2 cell lines and increases their susceptibility to NK lysis. Cancer Immunol Immunother (2009). 441. Diermayr, S., et al. NKG2D ligand expression in AML increases in response to HDAC inhibitor valproic acid and contributes to allorecognition by NK-cell lines with single KIR-HLA class I specificities. Blood 111, 1428-1436 (2008). 442. Skov, S., et al. Cancer cells become susceptible to natural killer cell killing after exposure to histone deacetylase inhibitors due to glycogen synthase kinase-3- dependent expression of MHC class I-related chain A and B. Cancer Res 65, 11136-11145 (2005). 443. Armeanu, S., et al. Natural killer cell-mediated lysis of hepatoma cells via specific induction of NKG2D ligands by the histone deacetylase inhibitor sodium valproate. Cancer Res 65, 6321-6329 (2005). 444. Ogbomo, H., Michaelis, M., Kreuter, J., Doerr, H.W. & Cinatl, J. Histone deacetylase inhibitors suppress natural killer cell cytolytic activity. FEBS Lett 581, 1317-1322 (2007). 445. Lundberg, P., et al. The immune response to herpes simplex virus type 1 infection in susceptible mice is a major cause of central nervous system pathology resulting in fatal encephalitis. J Virol 82, 7078-7088 (2008). 446. Singh, S.K., et al. Identification of human brain tumour initiating cells. Nature 432, 396-401 (2004). 447. Pek, E.A., Chan, T., Reid, S. & Ashkar, A.A. Characterization and IL-15 dependence of NK cells in humanized mice. Immunobiology (2010). 448. Gousias, K., et al. Frequent abnormalities of the immune system in gliomas and correlation with the WHO grading system of malignancy. J Neuroimmunol (2010). 449. Boutell, C., et al. Herpes simplex virus type 1 ICP0 phosphorylation mutants impair the E3 ubiquitin ligase activity of ICP0 in a cell type-dependent manner. J Virol 82, 10647-10656 (2008). 450. Halford, W.P., Puschel, R. & Rakowski, B. Herpes simplex virus 2 ICP0 mutant viruses are avirulent and immunogenic: implications for a genital herpes vaccine. PLoS One 5, e12251. 451. Johnson, K.E. & Knipe, D.M. Herpes simplex virus-1 infection causes the secretion of a type I interferon-antagonizing protein and inhibits signaling at or before Jak-1 activation. Virology 396, 21-29. 452. Paladino, P. & Mossman, K.L. Mechanisms employed by herpes simplex virus 1 to inhibit the interferon response. J Interferon Cytokine Res 29, 599-607 (2009). 453. Huard, B. & Früh, K. A role for MHC class I down-regulation in NK cell lysis of herpes virus-infected cells. Eur J Immunol 30, 509-515 (2000).

215