1 Characterization of Wwox Expression and Function in Canine

Characterization of Wwox Expression and Function in Canine Mast Cell Tumors and

Malignant Mast Cell Lines

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in

the Graduate School of The Ohio State University

Rebecca Lynn Makii

Graduate Program in Comparative and Veterinary Medicine

The Ohio State University

2020

Master’s Examination Committee

Joelle Fenger, DVM, PhD, DACVIM, Advisor

Eric Green, DVM, DACVR

Ryan Jennings, DVM, PhD, DACVP

Copyrighted by

Rebecca Lynn Makii

2020

Abstract

Mast cell tumors (MCT) are the most common skin tumor in dogs with behavior varying from benign to aggressive, metastatic disease. While activating mutations in the receptor tyrosine kinase KIT (c-KIT) have been identified in up to 30% of high-grade

MCTs, the genetic alterations driving tumorigenesis in the 70% of MCTs that do not possess c-KIT mutations remains unclear. The WW domain-containing oxidoreductase

(WWOX) tumor suppressor gene is frequently lost or attenuated in many human cancers and cancer cell lines and data suggest that loss of WWOX impedes DNA damage response (DDR) and repair leading to genomic instability. The purpose of this study was to characterize WWOX expression in spontaneous canine mast MCTs and mastocytoma cell lines and begin to define the functional consequences of WWOX inactivation on mast cell viability and clonogenic survival in response to double-stranded DNA (dsDNA) damaging agents. qRT-PCR and Western blotting showed that WWOX is decreased in

MC lines and primary MCTs compared to bone marrow-cultured MCs, suggesting that loss of WWOX is a frequent event in this disease. WWOX expression was assessed by immunohistochemistry in paired normal dermal MCs (N = 15), low-grade MCTs (N =

14), and high-grade MCTs (N = 5) and demonstrated that there is decreased percent of cells staining for Wwox in high-grade MCTs. To better define the functional consequences of WWOX loss on MC behavior, MCs transduced with control or WWOX ii lentiviral or sh-RNAs targeting WWOX were treated with ionizing radiation, and cell survival and viability were assessed by clonogenicity and MTT assays. Overexpression of

WWox in the BR MC line did not alter DDR or cell viability; however, further decreasing expression of WWOX in the C2 MC line conferred a survival advantage post- irradiation. Lastly, we demonstrate validation of tissue specific WWOX knockout a mouse model to better understand the role of WWOX in normal mast cells. These findings provide insight into the functions of WWOX in MCs with the ultimate goal of identifying novel targets for therapeutic intervention.

iii

Dedication

In loving memory of Patricia & Chester Berlin and Mildred & Melvin Makii

Dedicated to my parents, Michael and Christine Makii and my siblings, Jason Makii,

Justin Makii, and Amanda Fee, for their constant love, support, and encouragement throughout the years. To my nieces and nephews: I hope you fall in love with science just as much as I have.

Acknowledgments

I would like to begin by expressing my sincerest gratitude to my mentor, Dr.

Joelle Fenger, for her patience, guidance, support, and encouragement through both my graduate and veterinary programs. Your willingness to take me in as a “lost” graduate student developed into an invaluable opportunity to learn how to ask questions, think critically, and gain confidence within the field of comparative medicine.

The efforts of this work would not have been possible without the friendship, technical support, and morale encouragement of current and former laboratory members,

Dr. Justin Breitbach, Hanna Cook, Feng Xu, and Darian Louke. I would additionally like to thank Wessel Dirksen and Dr. Tom Rosol for providing me with foundational laboratory training during my graduate programming.

I am also appreciative of the guidance and technical assistance from my thesis committee, Drs. Eric Green and Ryan Jennings. Your expertise and guidance in this project has provided me with greater insight into both radiation oncology and pathology and an admiration for the collaborative nature of the veterinary profession.

Additional technical help for this project was provided through Dr. Sue

Knoblaugh, Chelssie Breece, and David Hart from the Comparative Pathology and

Mouse Phenotyping Shared Resource, Dr. Chris Premanandan and Jenny Bolon from the

Applied Pathology Services, Dr. Noopur Desai, Dr. Jillian Walz, and Nicole Pasternak of v the Integrated Oncology Service, Dr. Holly Borghese of the Blue Buffalo Clinical Trials

Office Tissue Bank, and Alex Cornwell of the Analytical Cytometry Shared Resource.

Each of you played an instrumental role in ensuring the completion of this project through your technical expertise, encouragement, and troubleshooting.

To my veterinary classmates (particularly Lindsay Courtney, Leah Giralico,

Megan Goodnight, Kelsey Bick, Sarah Bek Komnenovich, Kristen Behrens, Jessica

Battyanyi, and Andrea Bessler), the fourth floor of VMAB, and the pathology and oncology residents that I have befriended in my graduate studies, thank you for the camaraderie, support, and friendship over the years. I sincerely cannot wait to see the exciting pathways each one of us take as we embark further into our careers.

Finally, with the most gratitude, I would like to thank my biggest supporter,

Michael Kemp, my dearest friends (Serena Nayee, Abby McClaine, Akul Yajnik, Andy

Goldfarb, Megan Huber, Brady Schoeffler and Emily Selio), and my family (the Makiis, the Fees, the Kemps, and the Martins) for their encouragement, understanding, patience, and love. You all mean more to me than you know.

Vita

2012 ………………………………………………………..Mount de Sales Academy

2016 ………………………………………………………..B.S. Pharmaceutical Sciences,

The Ohio State University

2016 – present ……………………………………………..M.S./D.V.M. candidate

The Ohio State University

Fields of Study

Major Field: Comparative Veterinary Medicine

vii

Table of Contents

Abstract ...... ii

Dedication ...... iv

Acknowledgments ...... v

Vita ...... vii

List of Tables ...... xi

List of Figures ...... xii

Chapter 1. Literature Review ...... 1

1.1 The WW-Domain Containing Oxidoreductase (WWOX) Gene ...... 1

1.2 In Vivo Functions of WWOX ...... 3

1.3 WWOX Dysregulation in Cancer ...... 5

1.4 WWOX Functions in DNA Damage Repair and Genomic Stability ...... 8

1.5 Canine Mast Cell Tumors ...... 10

1.6 Summary ...... 13

Chapter 2: Characterizing the Role of WWOX in Canine Mast Cell Tumors and Cell

Lines ...... 15

2.1 Abstract ...... 15

2.2 Introduction ...... 16 viii

2.3 Materials and Methods ...... 20

2.3.1 Tumor microarray construction and immunohistochemistry ...... 20

2.3.2 Cell lines and reagents ...... 21

2.3.3 RNA isolation, cDNA synthesis, and qRT- PCR ...... 22

2.3.4 Immunoblotting ...... 23

2.3.5 Recombinant stem cell factor co-culture ...... 24

2.3.6 WWOX and shWWOX lentivirus infection ...... 24

2.3.8 Assessment of cell viability and cell proliferation ...... 26

2.3.9 Clonogenic survival assay ...... 27

2.3.11 Generation of CPA3-Cre;WWOXfl/fl transgenic mouse model ...... 27

2.3.13 Statistics ...... 28

2.4 Results ...... 28

2.4.1 Wwox expression is frequently decreased in primary canine MCTs and canine

MC lines ...... 28

2.4.2 Loss of Wwox expression is associated with increased tumor grade in primary

canine MCTs ...... 29

2.4.3 WWOX overexpression does not affect cell viability or clonogenic survival

following IR in the canine BR mast cell lines ...... 31

2.4.4 WWOX knockout affects clonogenic survival following ionizing radiation in

the canine C2 mast cell line and this is independent of cell viability ...... 31 ix

2.4.5 Characterization of CPA3-Cre;WWOXfl/fl transgenic mouse model ...... 32

2.5 Discussion ...... 32

2.6 Conclusions & Future Directions ...... 36

Appendix: Tables and Figures ...... 38

References ...... 55

List of Tables

Table 1: Scoring criteria for tumor microarray ...... 38

Table 2: Primers for quantitative reverse transcriptase polymerase chain reaction ...... 39

Table 3: Density of seeding for clonogenic assay ...... 40

List of Figures

Figure 1: Wwox transcript is decreased in primary canine MCTs and MC lines ...... 41

Figure 2: Wwox protein expression in primary canine MCTs and MC lines ...... 42

Figure 3: Loss of Wwox expression is associated with increased histologic tumor grade in primary canine MCTs ...... 43

Figure 4: Generation of canine MC lines expressing Wwox lentiviral constructs ...... 45

Figure 5: Generation of canine MC lines expressing WWOX-targeted shRNA lentiviral constructs ...... 47

Figure 6: Alteration of Wwox does not affect cell viability in BR or C2 mast cells ...... 49

Figure 7: Wwox knockout affects clonogenic survival following ionizing radiation in the canine C2 mast cell line ...... 51

Figure 8: CPA3-Cre;WWOXfl/fl-Tg mouse model validation ...... 53

Figure 9: Influence of SCF Supplementation ...... 54

xii

Chapter 1. Literature Review

1.1 The WW-Domain Containing Oxidoreductase (WWOX) Gene

The WW-domain containing oxidoreductase (WWOX) gene is a putative tumor suppressor gene that spans a large genomic region on the long arm of human chromosome 16 at band q23 approximately. Prior to the discovery, mapping and cloning of WWOX in 2000, the term “fragile site” was used to describe recurrent chromosomal perturbations in the form of gaps, breaks, and rearrangements at specific DNA loci on metaphase chromosomes [1, 2]. Early cytogenetic data led to the classification of chromosomal fragile sites as common (CFSs) or rare, depending on the frequency in the population as well as their inheritance pattern. It was later concluded that the genomic region in which WWOX resides is the same as that of the second most common human chromosomal fragile site, FRA16D at 16q23 [1]. The intriguing discovery that CFSs are late replicating regions known to be preferential hotspots for metaphase chromosome breaks and rearrangements led to scientific efforts to investigate the biological functions of WWOX and study its relevance in human health and disease.

The WWOX gene spans over 1 megabases (Mb) of the genomic locus and encodes a 1.245 kilobase (kb) open reading frame containing 9 exons separated by characteristically large introns, common in genes spanning CFSs. WWOX is translated

1 into a 414-amino acid (46 kDa) WWOX protein that localizes primarily in the cytoplasm near the Golgi apparatus [3]. The WWOX protein, as the name suggests, contains two N- terminal WW-domains and a central region homologous to the short-chain dehydrogenase/reductase (SDR) family. The first WW-domain (WW1) is involved in protein-protein interactions by binding to partner proteins harboring proline-rich PPxY motifs and acts as an adaptor protein regulating their localization, transactivation, and stability [4–6]. The SDR region functions through NAD(H) and NADP(H) co-factors to catalyze reactions involving a variety of substrates and contains a catalytic signature motif conserved in steroid dehydrogenases, suggesting that it is likely to be involved in steroid metabolism [7]. Comprehensive analyses of WWOX protein expression in non- neoplastic human tissues demonstrate that WWOX is globally expressed in a variety of cell types and is preferentially highly expressed in secretory epithelial cells of reproductive, endocrine and exocrine organs (breast, ovaries, prostate, and testes) [8].

Interestingly, minimal to no expression of WWOX was detected in adipose, connective, and lymphoid tissues, myelinized structures and blood vessels; however, the significance of tissue-specific WWOX expression is unknown.

Phylogenetic analysis of WWOX protein suggest that WWOX is an evolutionarily conserved protein. Although all species exhibit distinct sequence relationships of WWOX orthologues through evolution, the conserved sequences of WW domains and SDR domain in WWOX imply that it is tightly regulated across species. The WWOX gene is remarkably well conserved in sequence for example, between human WWOX and canine

WWOX both at exon as well as intron levels exhibit 91.2% identity along with 94.4% similarity at the protein level [9].

1.2 In Vivo Functions of WWOX

To begin to understand the biological functions of WWOX both at the molecular and cellular level, several murine models of WWOX ablation were established and studied. Aqeilan et al. established the first knockout (KO) mouse model for Wwox in

2007 [10]. Using homologous recombination, the authors altered the genomic locus of murine Wwox by replacing exons 2-4 of Wwox with a targeting cassette thus generating a

Wwox mutant allele. While wild type (WT), heterozygous (Wwox +/-) and Wwox -/- mice appeared normal in size at birth, Wwox -/- mice showed post-natal growth retardation and premature death at 3-4 weeks of age. These KO pups displayed severe metabolic defects associated with hypoglycemia, hypolipidemia, hypocalcemia, as well as impaired expression of steroidogenic enzymes suggesting a critical role for WWOX in cellular metabolism [11]. Wwox null mice exhibited deficiencies in steroidogenesis resulting in a phenotypic presentation of failed Leydig cell development and reduced theca cell proliferation [11]. Concordant with the severe metabolic defects observed in KO mice,

Wwox ablation led to impaired bone metabolism, which manifested as decreased trabecular bone density (osteopenic phenotype) and delayed bone formation. To this end, ex vivo calvarial osteoblast cultures generated from Wwox -/- mice demonstrate that the delay in bone formation likely results from a cell autonomous defect in osteoblast differentiation beginning at the mineralization stage. RUNX2, the principal 3 transcriptional regulator of osteoblast differentiation, was found to physically associate with WWOX and suppress its transactivation ability in osteoblast [12]

To overcome the limitations in early post-natal lethality in Wwox -/- mice, conditional Wwox-knockout mice (Wwoxfl/fl) were established and crossed with EIIA-Cre general deleter mice to generate Wwox KO (WwoxΔ/Δ) mice in all tissues [13]. In agreement with the observed phenotype in prior Wwox KO models, WwoxΔ/Δ mice displayed significant skeletal growth impairments and 100% of mice died by 3-4 weeks of age as a consequence of hypoglycemia and a severe metabolic disorder. Recent studies provide evidence that WWOX functions in regulating glycolysis via its interactions with

HIF1-a and suggest that WWOX dysregulation results in altered cellular metabolism

[14]. WWOX expression is also dependent on the metabolic setting; WWOX levels increase during oxidative phosphorylation and decrease during hypoxic conditions [15].

Lastly, data generated in animal models show that WWOX interacts with the mitochondrial proteins, IDH and SOD1, implying that during oxidative phosphorylation,

WWOX plays a role in regulating reactive oxygen species handling and maintaining the stability of genetic material [16].

Subsequent studies investigating the in vivo consequences of Wwox inactivation observed that Wwox null mice exhibit spontaneous seizure activity and signs of neurodegeneration [17]. Wwox gene mutations have been associated with epilepsy and ataxia in mouse and rat animal models and numerous reports have documented WWOX germline mutations in humans that are associated with developmental retardation, ataxia, and early onset of epilepsy [18, 19]. Collectively, data generated from animal models of

Wwox inactivation support a critical role for Wwox in early post-natal development, growth and normal cellular metabolism.

1.3 WWOX Dysregulation in Cancer

WWOX was identified as a putative tumor suppressor in human cancer cells because it spans the entire region of the second most common chromosomal fragile site

(CFS), FRA16D (16q23.3–24.1) [1]. Interestingly, orthologs of WWOX show a high degree of interspecies conservation and in mice, the genomic region surrounding Wwox is also known as CFS, Fra8E1[20]. In domestic dogs, WWOX maps to chromosome 5; however, the location of CFSs in the canine genome and the relative frequency of CFSs in the canine population is not well characterized. Among vertebrates, CFSs represent genomic areas prone to chromosome breakage during metaphase after inhibition of DNA replication by aphidicolin, an alpha-DNA polymerase inhibitor, or using folate deprivation [21]. Given that CFSs are high susceptible to chromosomal breaks, CFSs are thought to be involved in chromosomal rearrangements observed in cancer [22]

Early studies demonstrated that FRA16D is frequently deleted or subject to nonrandom translocation events in human breast cancer cells [23]. This region is commonly lost in ductal carcinoma in situ of the breast and prostate cancer, suggesting that genes encompassing this CFS function as tumor suppressors in the early stages of tumorigenesis [24, 25]. In vivo studies investigating the phenotypic consequences of

Wwox deletion found that the incidence of spontaneous tumor formation in increased in

Wwox KO (Wwox -/-) mice in comparison to wild-type or heterozygous littermate mice 5

[10]. Additionally, Wwox (+/-) mice developed significantly more ethyl nitrosurea- induced lung tumors and lymphoma compared to wild-type littermate mice, suggesting that Wwox haploinsufficiency is cancer predisposing. In a mammary tumor-susceptible mouse model, 50% of heterozygous Wwox (+/-) females developed breast cancer compared to 7% in the control group. These data suggest that breast cancer cells with

LOH of WWOX may require additional genetic alterations for malignant transformation

[26].

Consistent with these findings, loss of WWOX occurs in a number of human malignancies including bone, lung, liver, prostate, ovarian, and pancreatic cancers, among others [27]. Loss of WWOX is thought to occur predominantly through loss of heterozygosity (LOH) or homozygous deletion of the gene [28]. It is likely, however, that a number of other epigenetic and genetic factors play a role in silencing WWOX expression. For example, hypermethylation of CpG islands at the WWOX promoter transcriptionally inactivates the tumor suppressor WWOX in ovarian, pancreatic, lung, and breast cancer cell lines [29–33]. In contrast, the MCF7 breast cancer cell line does not show evidence of CpG methylation in the promoter region of WWOX [34]. In this cell line, aberrant splicing events result in atypical nuclear localization of WWOX and attenuated function.

WWOX has been suggested to play roles in multiple cellular processes including but not limited to: cellular metabolism, cell cycle checkpoint activation, and maintaining genomic stability. The versatile nature of WWOX is attributed, in part, to its ability to interact with different proteins in multiple cellular pathways. One of the first binding

6 partners identified to interact with WWOX was p73, a p53 family homolog that mediates cellular response to stress, including regulating cell cycle checkpoint activation and apoptosis [35]. The direct interaction of WWOX with its binding partners results in cytoplasmic sequestration and inactivation [5].

For example, attenuated or complete loss of WWOX is common in human breast cancer tissues and cell lines and deletion of FRA16D occurs in up to 80% of breast cancer tissues [24]. To date, several studies have identified WWOX interactors that mediate the aggressive biological behavior of breast cancer cells in vitro. ErbB4, a receptor tyrosine kinase that promotes cell proliferation and survival of mammary epithelial cells was identified as a protein partner of WWOX [36]. To determine whether

WWOX plays a tumor suppressor role, human breast cancer cell lines expressing low basal levels of WWOX were transfected with WWOX overexpression constructs.

WWOX reconstitution decreased soft agar colony formation and significantly decreased breast cancer tumor growth in vivo in mouse xenograft tumor models, in part, through inhibition of ErbB4 signaling [34].

Juvenile Wwox null mice display a higher incidence of spontaneous osteosarcoma development implicating WWOX dysregulation in the pathogenesis of this disease. Prior studies found that WWOX deletions are present in 30% of human OS tumors.

Interestingly, WWOX protein expression was reduced or absent in 60% of tumors, suggesting that epigenetic and genetic factors likely influence WWOX expression in OS cells [37, 38]. Overexpression of WWOX in human OS cells resulted in decreased invasion, anchorage independent cell growth, colony formation, and tumorigenicity in

7 nude mice [38]. In OS cell lines, these effects were mediated through interactions with various proteins such as RUNX2 metastatic target genes [39].

1.4 WWOX Functions in DNA Damage Repair and Genomic Stability

The capacity to maintain genomic stability is reliant upon a cell to recognize and sense DNA damage and successfully execute downstream pathways to repair DNA breaks. Early evidence that WWOX plays a critical role in sensing and repairing DNA damage came from the observation that WWOX transcript and protein expression transiently increased in response to ionizing radiation (a potent inducer of double- stranded DNA breaks) exposure in MCF7 breast cancer cells. Abu-Odeh et al. identified a mechanism by which WWOX interacts with ataxia telangiectasia-mutated (ATM) to mediate the DNA damage repair (DDR) response after induction of double-stranded

DNA breaks (DSB) [40]. In WWOX-deficient MCF7 cells, ATM activation at Ser1981- phosphorylated-(p-)-ATM was diminished and accompanied by reduced phosphorylation of downstream ATM substrates KAP1 and H2AX. Reconstitution of wild type WWOX in these cells restored ATM phosphorylation and rescued the activation of ATM substrates (H2AX, KAP1, p53, and CHK2) responsible for controlling cell cycle arrest, genomic repair, and/or apoptosis [41, 42]. Importantly, this study demonstrated that shortly after the induction of DNA damage, a series of post-translational modifications

(primarily polyubiquitination by the ubiquitin E2 ligase ITCH) allows for WWOX accumulation and localization in the nucleus where it complexes with ATM [40]. In

WWOX-deficient MCF7 cells, loss of WWOX resulted in an increase in DSBs and chromosomal instability as assessed by standard comet assays.

Similarly, WWOX functions in regulating single-stranded DNA damage repair via interaction with Ataxia telangiectasia and Rad3-related protein (ATR). When single- stranded DNA breaks (SSBs) are sensed by ATR, it localizes at the replication fork and recruits other proteins, such as H2AX and CHK1, to stall replication and initiate repair mechanisms [42]. In cell lines, treatment with the chemical agents aphidicolin or hydroxyurea or exposure to UV radiation induces SSBs and increases the expression of

WWOX protein [43]. Reciprocal experiments demonstrated that WWOX depletion in cell lines reduced phosphorylation and activation of H2AX, CHK1 and increased chromosomal breaks; WWOX restoration rescued this phenotype. WWOX also co- localizes with ATM upon single-stranded DNA damage, suggesting that this complexation is required prior to activation of the ATR checkpoint pathway. These data suggest that WWOX influences ATR-mediated checkpoint activation in an ATM- dependent manner to initiate the DNA damage repair response following SSBs [43].

Moreover, these findings support a “caretaker” role for WWOX in the maintenance genomic stability through both ATM and ATR-mediated signaling. As such, WWOX attenuation or loss in cancer cells may contribute to defective DDR signaling, thereby leading to wide-spread genomic instability and promoting tumorigenesis.

1.5 Canine Mast Cell Tumors

Mast cell tumors (MCTs) are the most frequent cutaneous tumor of the dog [44–

47]. The clinical behavior of MCTs can vary with some dogs presenting with relatively benign MCTs where surgical excision is curative, to biologically aggressive, metastatic

MCTs that are refractory to chemotherapy, radiation therapy, and surgery. Despite aggressive, multimodal therapy for metastatic MCTs, up to 90% of dogs are ultimately euthanized due to complications associated with mast cell degranulation, such as anaphylaxis, gastrointestinal ulceration and/or perforation and mean survival times rarely exceed 6-9 months [48]. While the clinical diagnosis of malignant mast cell disease in dogs is relatively easy with standard cytologic evaluation, differentiating between those dogs predicted to have a benign or aggressive disease course can be quite challenging

[49]. Nearly 30% of dogs diagnosed with low-grade (grade I or II) MCTs go on to develop additional tumors of either primary or metastatic origin, emphasizing the need for better prognostic indicators [48].

The predominant method of clinical outcome prediction for MCTs has relied heavily on histological grading. For the past 30 years, the three-tier method developed by

Patnaik (grades 1, 2, 3, with grade 3 as most malignant) has been the most utilized grading scheme to provide prognostication and guide treatment decision-making for

MCTs [50]. This grading system focuses on histological characteristics such as cellularity, morphology, mitotic figures, stromal reaction, and invasiveness. However, the

Patnaik grading system has been criticized due to being largely subjective in nature with high inter-observer variability [51]. Additionally, the Patnaik system has shown to offer 10 little prognostic value in nearly 50% of sampled MCTs, predominantly due to the ambiguous categorization of grade 2 tumors when compared to grade 1 [48]. In fact, up to 15% of dogs with reportedly low-grade cutaneous mast cell tumors have regional lymph node spread upon presentation [52]. A new two-tier scheme (high- and low-grade) has been proposed by Kiupel and is predominantly based on number of mitotic figures and nuclear characteristics such as karyomegaly, multinucleation (defined as three or more nuclei), and atypia [48]. Additionally, utilization of this system has provided less speculation for prognosis, with low-grade median survival time (MST) surpassing 2 years compared to the high-grade MST of 4 months. This new system has improved consistency in grading and prognosis considerably, however, both systems are currently utilized at many institutions. Despite this improvement in consistency, nearly 20% of tumors identified as low-grade Kiupel and intermediate grade Patnaik will have metastatic disease, rendering histologic grading alone as not reliable enough to guide treatment decisions [52]. Additional biochemical methods of differentiating aggressive

MCTs from benign, such as Ki-67 and surviving immunohistochemistry, have been explored, however they have resulted in little prognostic value [53].

The increased incidence of MCTs in certain canine breeds suggest a strong genetic component guiding malignant transformation. The breeds reported to be at highest risk for the development of MCTs include Shar-Peis, American Staffordshire

Terriers, Labrador Retrievers, French Bulldogs, and Boxers [54, 55]; however, the causative genes that contribute to the development of MCTs in these dog breeds remains unclear. The discovery of somatic mutations within the proto-oncogene c-KIT in dogs

11 with high-grade MCTs has provided some insight into the signaling pathways that drive tumorigenesis. Activating mutations in the tyrosine kinase growth factor receptor KIT

(c-KIT) are present in approximately 25-30% of all canine MCTs, including up to 50% of high-grade tumors [56, 57]. Their discovery has led to the development of small molecule inhibitors of KIT that exhibit substantial biological activity in MCTs harboring wild-type or mutant KIT [58, 59]. Although the frequency of c-KIT mutations is increased in high grade MCTs, the other genomic alterations underlying the aggressive biological behavior of MCTs that do not possess c-KIT mutations remains unclear.

More recent studies have begun to shed light on the genomic landscape of canine

MCTs. Blacklock identified a unique gene expression profile that discriminated between biologically low grade, non-metastatic and biologically aggressive, high grade metastatic

MCTs [60]. Several genes implicated in promoting the metastatic phenotype were differentially expressed in non-metastatic MCTs compared to metastatic MCTs, including decreased expression of cellular adhesions proteins, SCRIB and DSP. Other genomic studies identified a 13-gene signature composed of genetic markers involved in cell cycle, DNA replication, and DNA repair that accurately discriminated between well- differentiated and poorly differentiated MCTs tumors; however, functional studies are necessary to determine the influence of these genes in malignant mast cell behavior [61].

Genome-wide DNA copy number analysis of canine MCTs demonstrate that

MCTs show a broad range in genome-wide copy number aberrations (CNA) and the frequency of CNAs increases with increasing histological grade [62]. Additionally, four

CNAs associated with high-risk MCTs were identified, including genomic loss of CFA 5

(-37.8 Mb) a region harboring the canine WWOX allele. These data support the notion that high-risk canine MCTs have an increased frequency of genome-wide DNA copy number aberrations and suggest that loss of genes, such as WWOX, within copy number aberrant regions may contribute to the development and progression of canine MCTs.

Currently there is limited information regarding the potential role of WWOX dysregulation in spontaneous canine MCTs. Given the known functions of WWOX in regulating DNA damage repair responses, a detailed understanding of the function of

WWOX in influencing multiple aspects of canine mast cell biology would be ideal to more accurately inform future clinical studies.

1.6 Summary

WW-domain containing oxidoreductase (WWOX) is tumor suppressor gene that is frequently attenuated or lost in a number of human cancers and tumor-derived cell lines.

While the loss of WWOX is associated with an increased incidence of spontaneous tumor development in transgenic mouse models, there is limited data regarding the potential role of WWOX in spontaneous canine malignancies. The purpose of this work was to investigate the potential role of WWOX dysregulation in the biology of canine mast cell tumors (MCTs) and mastocytoma cell lines. The first aim of this study was to evaluate

WWOX expression in low grade and high grade canine MCTs, malignant canine mast cell lines, and normal canine bone-marrow-cultured mast cells (BMCMCs). The next aim was to define the functional consequences of altered WWOX expression on DNA damage response and repair pathways in canine mast cell lines. Based on our findings 13 that WWOX expression is decreased in canine MCTs and that WWOX inactivation enhances viability and clonogenic survival of canine mast cell lines, the last objective was to generate a transgenic mouse model that allows for conditional deletion of WWOX in mast cells and basophils and validate that this mouse model functions in a tissue- specific manner. In summary, data generated from this project provides insight into the role of WWOX in normal and malignant mast cell biology. We have established a mouse model to aid in the study of WWOX biology and characterize the mechanisms by which

WWOX influences the behavior of mast cells in vitro and in vivo.

Chapter 2: Characterizing the Role of WWOX in Canine Mast Cell Tumors and Cell

Lines

2.1 Abstract

MCTs, the genetic alterations driving tumorigenesis in the 70% of MCTs that do not possess c-KIT mutations remains unclear. The WW domain-containing oxidoreductase

15 lentiviral or sh-RNAs targeting WWOX were treated with ionizing radiation, and cell survival and viability were assessed by clonogenicity and MTT assays. Overexpression of

2.2 Introduction

Mast cell tumors (MCTs) are the most common malignant skin tumor in dogs, accounting for approximately 7-20% of all cutaneous tumors [44–47, 63]. The biological behavior of canine MCTs is extremely variable, ranging from solitary, benign tumors to extremely aggressive tumors that metastasize to locoregional lymph nodes and distant organ sites. Several prognostic factors, including clinical stage and histopathologic grade using a 3- or 2-tier grading system aid in the classification of canine MCTs; however, the prognostic significance of histologic grade is associated with survival time and may not accurately predict metastasis. To this end, 37.5% of MCTs classified as ‘low grade’ using the Kiupel histologic grading system were from dogs with distant metastatic disease and

21.9% of ‘high grade’ MCTs were from dogs without evidence of distant metastases [48].

The etiology of canine MCTs is largely unknown; however, the identification of activating mutations in the proto-oncogene c-KIT in approximately 30% of dogs with 16 aggressive MCTs has provided insight into the genetic changes that mediate the biological behavior of MCTs [56, 57, 64, 65]. It has also resulted in the successful development and approval of a novel targeted therapeutic, Toceranib phosphate

(Palladia®) that works primarily by inhibiting KIT signaling [58]. While data suggests that KIT inhibitors have significant biologic efficacy in the setting of KIT mutation, responses are generally not durable beyond 12 months and treatment is often unsuccessful in the ~70% of dogs that do not possess KIT mutations [56, 57]. While the role of KIT dysfunction in mast cell neoplasia has been well described, a more complete understanding of the additional molecular factors that influence malignant mast cell behavior is necessary to identify novel targets and to develop new therapeutic approaches for MCTs. To this end, recent genome-wide gene expression analyses suggest that the presence of distinct subclasses of low- and high-risk MCTs exist with respect to their underlying molecular phenotypes and prognoses [60, 66]. These include enrichment of factors associated with proliferation pathways and overexpression of genes associated with the extracellular matrix that are linked to the activity of cancer-associated fibroblasts present in high-risk MCT stroma. Similarly, genome-wide DNA copy number analyses demonstrate that recurrent DNA copy number aberrations (CNAs) are associated with

KIT mutation status and high histological grade, suggesting that loss or gain of genes within copy number aberrant regions may contribute to the neoplastic transformation of mast cells [62].

The WW domain-containing oxidoreductase (WWOX) is a highly conserved,

46 kDa protein consisting of two N-terminal WW domains and a C-terminal short-chain

17 dehydrogenase/reductase domain [3]. The first WW-domain (WW1) is involved in protein-protein interactions by binding to partner proteins harboring proline-rich PPxY motifs and acts as an adaptor protein regulating their localization, transactivation, and stability, thereby influencing normal physiology and development [4–6]. Given its role in cellular metabolism, cell cycle checkpoint activation, and maintenance of genomic stability, WWOX has been implicated in cancer initiation and progression [10, 28, 34].

Initial evidence of the role of WWOX in tumor biology came from the discovery that the human WWOX gene resides in the genomic region FRA16D at 16q23, a common fragile site that is prone to frequent chromosome breakage [1, 23]. There is now substantial data supporting the role of WWOX in cancer biology, including the observation that loss or attenuation of WWOX occurs in a number of human malignancies including bone, lung, liver, ovarian, and pancreatic cancers [27]. The genomic location of WWOX makes the locus susceptible to loss of heterozygosity (LOH) or homozygous deletion of the gene, resulting in reduced gene expression [28]. Alternatively, other epigenetic and genetic factors play a role in silencing WWOX expression. For example, hypermethylation of

CpG islands at the WWOX promoter has been shown to transcriptionally inactivate

WWOX in ovarian, pancreatic, lung, and breast cancer cell lines [29–33]. In contrast, the

MCF7 breast cancer cell line does not show evidence of CpG methylation in the WWOX gene promoter and aberrant splicing events result in atypical nuclear localization and attenuated WWOX function [34].

Importantly, in vivo studies investigating the phenotypic consequences of global

Wwox deletion found an increased incidence of spontaneous tumor formation, suggesting

18 a critical role for WWOX in tumorigenesis [10]. Additionally, heterozygous Wwox (+/-) mice developed significantly more ethyl nitrosourea-induced lung tumors and lymphoma compared to wild-type littermate mice, suggesting that Wwox haploinsufficiency is cancer predisposing. The versatile nature of WWOX is attributed, in part, to its ability to directly interact with proteins in a variety of cellular processes, including cell differentiation and growth. One of the first binding partners identified to interact with

WWOX was p73, a p53 family homolog that mediates cellular response to stress, including regulating cell cycle checkpoint activation and apoptosis [35]. Loss of WWOX contributes to tumor progression and chemotherapeutic resistance through its interaction with several binding partners, including AP-2γ, STAT3, and ErbB4 [36, 67, 68]. As such, it represents a potentially relevant target for therapeutic intervention in many cancers.

It is well established that WWOX expression is altered in many human malignancies and that WWOX functions as a tumor suppressor gene through dysregulation of target binding partners [13]. Currently there is limited information regarding the potential role of WWOX dysregulation in malignant mast cell disease. As such, we sought to investigate the potential role of WWOX dysregulation in the biological behavior of canine MCTs. The purpose of this study was to characterize the expression of WWOX in canine primary MCT samples and malignant mast cell lines and to determine the functional consequences of altered WWOX expression in canine mast cell lines.

2.3 Materials and Methods

2.3.1 Tumor microarray construction and immunohistochemistry

Primary cutaneous mast cell tumor (MCT) tissue samples were collected from clinical cases that presented to The Ohio State University Veterinary Medical Center

(OSU-VMC). Consent for tissue collection was obtained from all owners in accordance with an approved IACUC protocol (2010A0015) and collected by the OSU-VMC Blue

Buffalo Clinical Trials Office and Veterinary Clinical Research Shared Resource. A total of 26 primary MCTs with 19 paired normal skin biopsies were identified. Surgical and post-mortem collected tumor samples were placed in formalin and processed for routine paraffin embedding for histopathology. The medical records of all dogs were reviewed and data pertaining to signalment, staging, treatment, and survival were abstracted.

Representative areas of tumor tissue and skin containing vasculature and normal mast cells were identified on hematoxylin and eosin (HE) stained sections by a single, board- certified veterinary pathologist (RJ) and graded using the Patnaik and Kiupel grading schemes. 2.0 mm core samples were extracted from the corresponding areas on 45 paraffin embedded tissue blocks and inserted into predetermined sites on the tissue microarray (TMA) recipient block. Immunohistochemical staining was performed for

WWOX (Cat # NBP2-47579, Novus Biologicals, LLC, Centennial, CO, USA) on all

MCT and normal skin samples. Normal canine testes tissue served as a positive control.

Negative controls consisted of irrelevant isotype matched antibody at matched dilutions.

Both the construction of the TMA bloc and immunohistochemical staining were

20 performed by the OSU-CVM Comparative Pathology and Mouse Phenotyping Shared

Resource.

Slides were evaluated by light microscopy to assess WWOX immunoreactivity.

Overall WWOX signal intensity of normal and malignant mast cells was subjectively scored from 0 to 3 (0 = none to weak, 1 = mild, 2 = moderate, 3 = strong) by two boarded veterinary pathologists (RJ, CP) using the specifications detailed in Table 1. The percentage of positive cells was also estimated and scored from 0 to 4 (0 = no staining, 1

= 1-25%, 2 = 26-50%, 3 = 51-75%, 4 = >75%). Total scores were averaged between raters for comparison based on tumor grade.

2.3.2 Cell lines and reagents

The canine mastocytoma cell line BR (activating point mutation L575P in the JM domain of KIT) and C2 (KIT ITD mutation in the JM domain) were generously provided by Dr. Warren Gold (Cardiovascular Research Institute, University of California – San

Francisco, CA, USA) and murine cell lines P815 (D814V KIT mutation) and C57 (wild- type KIT) were generously provided by Dr. Stephen Galli (Stanford University, Stanford,

CA, USA). The cells were maintained in Roswell Park Memorial Institute medium

(RPMI) 1640 (Gibco® Life technologies, Grand Island, NY, USA), supplemented with

10% fetal bovine serum (FBS) (Catalog # 12483-020, Gibco® Life technologies),

Antibiotic-Antimycotic (Cat # 15240-062, Gibco® Life technologies), GlutaMAXTM (Cat

# 1963762, Gibco® Life technologies), non-essential amino acids, sodium pyruvate, and

HEPES (4-(2-dydroxethyl)-1-piperazineethanesulfonic acid) at 37ºC, supplemented with 21

5% CO2 (media supplements from Gibco®). Canine bone marrow cultured mast cells

(cBMCMCs) were generated from 2 dogs and maintained in Stemline (Sigma-Aldrich,

St. Louis, MO, USA) medium supplemented with recombinant canine stem cell factor (R

& D Systems, Minneapolis, MN, USA) as previously described [69].

2.3.3 RNA isolation, cDNA synthesis, and qRT- PCR

Total RNA was extracted from mast cell lines or primary MCTs using the RNeasy

Mini Kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. cDNA was made from 500 ng total RNA using Superscript III (Invitrogen ThermoFisher

Scientific, Waltham, MA, USA). Primers designed and utilized for canine WWOX and

18S endogenous control are listed in Table 2. An annealing temperature of 60 °C was used for all reactions. Standard PCR was performed with all primer sets and amplicon length was verified through agarose gel electrophoresis and visualization of PCR products using the Alpha Imager system (Alpha Innotech Corp, San Leandro, CA).

Real-time quantitative PCR (RT-qPCR) was performed to measure WWOX transcript expression using the methods described above. qRT-PCR was performed using

Applied Biosystem’s StepOne Plus Real-Time PCR system (Applied Biosystems

ThermoFisher Scientific). Canine WWOX and 18S were detected using Fast SYBR

Green PCR master mix according to the manufacturer’s instructions. Data normalization was performed relative to 18S internal control. Experiments were repeated three times using samples in triplicate and included non-template controls for each gene. Relative

22 gene expression for all RT-qPCR data was calculated using the comparative threshold cycle method [70].

2.3.4 Immunoblotting

To assess the relative expression of WWOX, protein lysates were generated from primary MCT tissues or mast cell lines. Briefly, frozen tumor samples were pulverized using a frozen mortar and pestle. The resulting powder was resuspended in liquid nitrogen, and transferred to 1.5 mL microcentrifuge tube. Tumor samples and mastocytoma cell lines were washed twice with 1X Dulbecco’s phosphate-buffered saline

(DPBS, Gibco) and resuspended in complete lysis buffer consisting of 20 mM Tris-HCl pH 8.0, 137 nM NaCl, 10% glycerol, 1% IPEGAL CA-630, 10 mM ethylenediaminetetraacetic acid (EDTA), 1 mg/mL aprotinin, 1 mg/mL pepstatin A, 1 mM phenylmethylsulphonyl fluoride, 1 mM sodium orthovanadate, and 10 mM sodium fluoride. Samples were rocked for 1 hour at 4°C, centrifuged for 15 minutes at 14,000

RPM at 4°C, and supernatants collected. Bradford protein quantification assay was performed on the extracts using BioRad Reagent (Cat #5000006, BioRad, Hercules, CA,

USA) according to manufacturer’s instructions. Equal amounts of total protein were separated on 12% SDS-PAGE gels (Cat #4561093, Bio-Rad) and transferred onto PVDF membranes. Membranes were blocked in TBS-T containing 3% non-fat dry milk for 1 hour and incubated overnight with polyclonal rabbit anti-WWOX antibody (Cat #PA5-

17237, Invitrogen ThermoFisher Scientific). The membranes were incubated with

23 appropriate horseradish peroxidase linked secondary antibody (Cat# 7074S, Cell

Signaling, Danvers, MA), washed, and exposed to luminol enhancer Supersignal® West

Dura Extended Duration Substrate developer (Cat #34075, ThermoFisher Scientific).

Blots were stripped using RestoreTM Western Blot Stripping Buffer (Cat #21059,

ThermoFisher Scientific), washed, and reprobed for β-actin (Cat #3700S, Cell Signaling,

Danvers, MA, USA). Band intensities were calculated using ImageJ software (NIH,

Bethesda, MD, USA) and relative intensity of WWOX was determined by dividing by β- actin.

2.3.5 Recombinant stem cell factor co-culture

To determine the potential effect of stem cell factor (SCF) stimulation on WWOX expression in canine mast cell lines, BR and C2 cells (1 x 107) were cultured in complete media supplemented with 100 ng/mL canine recombinant SCF (Cat #2278SC025, R&D

Systems Inc., Minneapolis, MN) or left untreated for 24 hours at 37ºC supplemented with

5% CO2. Cells were collected, washed twice with 1X DPBS, and immunoblotted to detect WWOX expression was performed as previously described.

2.3.6 WWOX and shWWOX lentivirus infection

Full-length canine WWOX cDNA was amplified by RT-PCR from normal canine testes and the resultant product was gel purified and sequenced. Canine WWOX cDNA

24 was ligated into the pGEMT plasmid vector (Promega, Madison, WI, USA), and subcloned into the pCDH-CMV-MSC-copGFP lentiviral expression plasmid (Cat

#CD511B-1, Systems Biosciences, Palo Alto, CA, USA). Packaging of the lentiviral constructs was performed using the pPACKH1 Lentivector Packaging Kit (Cat

#LV500A-1, System Biosciences) according to manufacturer’s instructions. Briefly, 1 x

106 BR cells were incubated overnight in complete medium. The following day, medium was changed to antibiotic-free medium with transfection agent TransDuxTM (Cat #

LV850A1, Systems Biosciences, Mountain View, CA, USA) and cells were infected with empty control (EV) lentivirus or WWOX lentivirus or TransDuxTM alone for 24 h.

FACS-mediated sorting based on GFP expression was performed 72 h post-transduction and WWOX overexpression was validated by qRT-PCR and Western blotting analysis.

Stable knock down of WWOX was performed using short hairpin RNA (shRNA) constructs cloned into the pGreenPuro shRNA cloning lentivector (Cat # SI505A-1,

System Biosciences) and high-titer lentiviral stocks were generated as described above.

Briefly, 1 x 104 C2 cells were plated and left overnight in complete medium. The following day, medium was replaced with serum-free medium and target cells were infected with transfection agent TransDuxTM (Systems Biosciences) and either pGreenPuro-Scramble or pGreenPuro-shWWOX virus or TransDuxTM alone for 24 h.

FACS-mediated sorting based on GFP expression was performed 72 h post-transduction.

Cells were collected and processed for qRT-PCR and Western blotting as described below to detect levels of WWOX and efficiency of knock down. Sequences of template canine DNA were as follows: pGreenPuro-shWWOX-554 (5’-

CCCGCTGTTCAAGCAGTTCTT-3’) and pGreenPuro-shWWOX-1304 (5’-

CTGCAGTATGACCTCCACTAC-3’).

2.3.8 Assessment of cell viability and cell proliferation

To determine the effects of WWOX dysregulation on mast cell viability, BR cells

(5 x 103) transduced with EV control or WWOX lentiviral constructs and C2 cells (5 x

103) expressing scramble , shWWOX-554 or shWWOX-1304 constructs were seeded in complete medium in quadruplicate in 96-well plates and incubated for 24, 48, 72 or 96 hours. To determine the influence of WWOX expression on mast cell viability following treatment with DNA damaging agents, BR cells (5 x 103) transduced with EV control or

WWOX lentiviral constructs were treated with increasing doses (0, 2, or 4 Gy) of ionizing radiation (IR) using a Siemans ONCOR linear accelerator and incubated for 24,

48, 72, or 96 hours. 10 µL of MTT reagent (Cat #11465007001, Roche, Mannheim,

Germany) was added to each well and cells were incubated for an additional 4 hours under normal culture conditions. Following incubation, 100 uL lysis buffer was added and the fluorescence was measured using a SpectraMax M2 microplate reader (Molecular

Devices, Sunnyvale, CA, USA) with excitation at 570 nm and emission 670 nm filters.

Cell viability and proliferation was calculated as the fold change of absorbance from 0 h control cells. Each experiment was repeated three times using samples in quadruplicate.

2.3.9 Clonogenic survival assay

BR cells expressing EV or WWOX constructs and C2 cells expressing scramble, shWWOX-554 or shWWOX-1304 constructs were treated with increasing doses (0, 2, 4, or 6 Gy) of ionizing radiation (Siemans ONCOR linear accelerator). Cells were counted and plated to appropriate density based on dose received (Table 3) and resuspended in

MethoCultTM (Cat #-4535, STEMCELL Technologies, Vancouver, BC) to a final volume of 1.1 mL per the manufacturer’s instructions. Cells were seeded in triplicate in a 6-well plate in standard culture conditions and the inter-well space supplemented by 8 mL sterile water to maintain humidity. Surviving colonies were counted 17 days post-IR using a counting grid, and the surviving fraction was determined per protocol previously described [71]. Experiments were performed in triplicate and repeated three times.

2.3.11 Generation of CPA3-Cre;WWOXfl/fl transgenic mouse model

Transgenic (Tg) mice carrying a conditional WWOX allele (WWOXfl/fl) were crossed with CPA3-Cre Tg mice that express Cre-recombinase under the carboxypeptidase A3 (CPA3) promoter that restricts expression to mast cells and basophils [72, 73]. Protein lysates were generated from lung, spleen, and liver tissues from CPA3-Cre (wild type, WT) and CPA3-Cre;WWOXfl/fl (knockout, KO) mice.

Primary mouse bone marrow-cultured mast cells (mBMCMCs) were differentiated in recombinant mouse IL-3 (50 ng/mL, R & D Systems, Minneapolis, MN, USA) for 4-6 weeks from WT, CPA3-Cre;WWOXfl/WT, and KO Tg mice. Protein lysates and RNA

27 were collected per previously described protocols and WWOX expression was validated via qRT-PCR and western blotting. Experiments were repeated with three separate litters to confirm in vivo system function.

2.3.13 Statistics

Whenever possible, experiments were performed in triplicate or quadruplicate and repeated three independent times. Data were presented as mean plus or minus standard deviation. qRT-PCR data was normalized to internal control (18S) and the ∆∆ Ct method was used to compare mRNA expression using Student’s t-test or one-way analysis of variance (ANOVA) [70]. TMA data was analyzed using Student’s t-test and inter-rater reliability was assessed via Cohen’s kappa with Landis & Koch interpretations [74].

Group comparisons in the MTT proliferation assays and clonogenic survival assays were analyzed by one-way ANOVA. Values of p < 0.05 were considered statistically significant.

2.4 Results

2.4.1 Wwox expression is frequently decreased in primary canine MCTs and canine MC lines

To characterize the expression of Wwox in primary MCTs and canine MC lines, we evaluated Wwox expression in primary MCT patient samples and canine (BR and C2)

28 malignant mast cell lines with canine BMCMCs and murine mastocytoma cell lines (C57 and P815) serving as positive controls. Real-time PCR data demonstrated that Wwox transcript expression is substantially decreased in primary canine MCTs as compared to canine BMCMCs. Though there was some variation in levels of Wwox expression among tumor samples, this may be attributed differences in the heterogeneity of tumor stroma, presence other inflammatory cells within the tumor microenvironment, and/or normal structures within the epidermis and dermis [8]. This data was supported by Wwox protein expression via western blot analysis. Real-time PCR data demonstrated that Wwox transcript expression is substantially decreased in malignant canine MC lines when compared to canine BMCMCs. The canine BR and C2 mast cell lines are derived from cutaneous tumors whereas the P815 murine cells is a leukemia of mast cell origin and the

C57 murine mast cells are of bone marrow origin [75–77]. Interestingly, differences in

Wwox transcript and protein expression were noted between the BR and C2 cell lines, with BR cells expressing significantly less Wwox transcript and protein compared to the

C2 cell line.

2.4.2 Loss of Wwox expression is associated with increased tumor grade in primary canine MCTs

To better demonstrate mast cell specific expression of Wwox, primary mast cell tumors and paired normal skin were immunostained for Wwox. Due to technical problems in TMA development, 31 of the original 45 samples were assessed in this study,

29 with an additional 9 samples currently in processing and 1 sample lost due to diminished quality. Normal mast cells in non-tumor samples were identified based on morphology and location in the perivascular region and described as having mild cytoplasmic staining. Additional background labeling in normal endothelium and epithelium was noted in both normal and tumor samples, and thus not counted in primary tumor assessment.

Degree of staining intensity was not found to correlate with histologic grade in the commercially available antibody but did correlate with our in-house antibody such that with high-grade tumors had significantly less stain intensity. The percent of neoplastic cells staining was found to be significantly different for both antibodies such that with high tumor grade, the percent of cells staining for Wwox decreased, which supports our hypothesis that loss of WWOX may correlate with a more aggressive tumor phenotype.

Weighted Kappa scores were determined using GraphPad Prism online calculator. For the

Novus antibody, there was fair interrater agreement for stain intensity (K = 0.393) and moderate agreement for percent of cells stained (K = 0.588). For the in-house antibody, there was fair inerrater agreement for both staining intensity and percent of cells stained

(K = 0.264 and 0.273, respectively).

2.4.3 WWOX overexpression does not affect cell viability or clonogenic survival following IR in the canine BR mast cell lines

To assess the effects of WWOX manipulation, canine BR cells expressing control vector (EV) or WWOX were assessed for differences in cell viability via MTT at multiple time points to indirectly assess proliferation. No differences in cell viability were noted indicating that restoration of WWOX does not impact cell viability or proliferation. To assess clonogenic survival following IR, cells were analyzed via standard colony formation assay to determine surviving fraction. BR cells did not grow in the methylcellulose-based media, so the impact of WWOX overexpression on clonogenic survival following exposure to IR was unable to be assessed accurately. Therefore, MTT assays were performed on BR cells to assess viability post-IR. No significant differences were noted between EV and WWOX groups, indicating that restoration of WWOX does not impact cell viability after IR exposure in this cell line.

2.4.4 WWOX knockout affects clonogenic survival following ionizing radiation in the canine C2 mast cell line and this is independent of cell viability

To assess the effects of WWOX manipulation, canine C2 cells expressing scramble control or shRNAs targeting WWOX (shWWOX-554, shWWOX-1304) were assessed for differences in cell viability via MTT at multiple time points to indirectly assess proliferation. No differences in cell viability were noted indicating that knockdown of WWOX does not impact cell viability or proliferation. To assess clonogenic survival

31 following IR, cells were analyzed via standard colony formation assay to determine surviving fraction. Cells expressing shRNAs targeting WWOX were noted to have increased clonogenic survival following radiation at 4 & 6 Gy. This suggests that loss of

WWOX confers a clonogenic survival advantage for cells after ionizing radiation exposure at therapeutically relevant levels.

2.4.5 Characterization of CPA3-Cre;WWOXfl/fl transgenic mouse model

To confirm the functionality of our transgenic mouse model, WWOX expression was assessed in mBMCMC cultures by real time PCR. CPA3-Cre;WWOXfl/fl and CPA3-

Cre;WWOXfl/WT mBMCMCs and were noted to have significantly less WWOX expression compared to wild type. Wwox expression was assessed in mBMCMC cultures and normal tissue (lung, spleen, liver) by western blot. CPA3-Cre;WWOXfl/fl mBMCMC cultures were noted to have no Wwox protein expression compared to wild type. For both mice, WWOX expression was observed in normal tissue (lung, spleen, and liver) confirming tissue specific knockout of WWOX.

2.5 Discussion

Recent data has correlated increased copy number aberrations with increasing histologic grade in canine mast cell tumors [62]. In tumors harboring cKIT mutations, deletion of CFA 5 and gain of CFA 31 were the most typical aberrations noted when

32 compared to wild-type KIT tumors. Interestingly, canine WWOX is located on CFA 5, however, the role of WWOX has not been explored in canine neoplasia or normal biology. The purpose of this study was to investigate the role that WWOX plays in normal and malignant mast cell biology.

We found that primary canine MCTs express low basal levels of both Wwox transcript and protein compared to that found in normal canine BMCMCs. Similar to primary MCTs, Wwox transcript and protein expression is decreased in malignant canine

MC lines compared to that found in normal canine BMCMCs. When coupled with the numerous reports of human neoplasia that also demonstrate frequent downregulation or loss of WWOX, this data supports WWOX as a tumor suppressor gene and suggests its loss may be involved in promoting malignant behavior in canine mast cells [5, 24, 25,

28].

Due to technical problems in processing, only 34 of the original 45 samples were assessed fully (Kiupel [Patnaik]: high-grade [III] n = 5, low-grade [II] n = 14; Normal n =

15). Though there is a decreased sample size of higher-grade tumors that were assessed compared to lower grade, the current results demonstrate an increased loss of Wwox staining with increasing histologic grade, which coupled with qRT-PCR and western blot data is likely attributed to loss of Wwox expression. An additional 9 samples (Kiupel

[Patnaik]: high-grade [III] n = 5, low-grade [II] n = 1; normal n = 3) are in processing to be scored in the aforementioned methods to help increase the sample size within the higher-grade category. The overall decreased access to higher-grade samples lay partially due to the increased prevalence of lower grade tumors, which account for up to 76% of

MCT cases [54]. Two samples were lost due to diminished sample quality (1 paired normal, 1 high-grade).

The current data was determined via visual pathologist scoring, yielding semiquantitative results. Current literature supports that this method of assessment is often adequate to yield correlation in terms of staining intensity but may hold some inter- and intra-observer variability in scoring the extent of staining [78, 79]. Interestingly, our weighted Kappa scores indicated greater agreement between pathologists in percent of cells staining rather than staining intensity. This may be due to differences in interpretation of the positive and negative staining controls that were provided. The addition of a third pathologists scores as well as the additional high-grade samples may help to better demonstrate a relationship. Additionally, pursuit of digital image analysis may yield higher quality results by increasing reproducibility [80].

To define the biological consequences of Wwox loss on mast cell behavior, canine BR and C2 MC lines were transduced with either control or WWOX constructs or scramble vector control or shRNA lentivirus targeting WWOX. We have validated stable overexpression or knockdown of Wwox via qRT-PCR and Western blotting in the canine

BR and C2 mast cell lines. Concordant with data generated in human cancer cell lines demonstrating that Wwox plays a critical role in regulating the DNA damage repair response (DDR), Wwox knockdown enhanced clonogenic survival in the canine C2 mast cell line following treatment with ionizing radiation, a common therapeutic for nonresectable, aggressive canine mast cell disease [40, 81]. This finding was independent

34 of cell viability and suggests that Wwox dysregulation may play a role in mediating the

DDR in this cell line in response to double-stranded DNA damage induced by IR.

WWOX was discovered to play a major role in regulation DNA damage repair response (DDR) induced by single- and double-stranded breaks (SSB and DSB, respectively) [40, 43]. This is mediated through direct interaction with the ATM and indirect interaction through the ATR pathways. Loss of WWOX was found to impair these pathway functions, causing DNA damage to accumulate and attenuation of cell- cycle checkpoint activation, thus affecting genomic stability. Cell models developed by

Abu-Odeh demonstrated that checkpoint activation restoration was achievable through reconstitution of WWOX.

Interestingly, overexpression of Wwox in the canine BR mast cell line does not alter cell viability following treatment with IR; however, this may be explained by differences in the expression of Wwox binding partners in distinct cell lines and/or cellular redundancy that influence the effects of Wwox function on cell line behavior.

Since there was decreased colony formation overall within the methylcellulose-based media in the empty vector and transduced BR cell lines, we cannot conclude the effects of WWOX restoration on clonogenic survival following IR. This finding may highlight the potential differences in cell surface proteins that exist between cell lines, further demonstrating the complicated genetic heterogeneity known in these tumors. Similar discrepancies in in vitro results were demonstrated with restoration of WWOX into human

PEO1 ovarian cancer cells, with in vivo studies revealing the true phenotype of gene

35 restoration [82]. Thus, functions of Wwox may be contextual in nature based on the molecular landscape of the cell and potential influence of the tumor microenvironment.

To better define the role of WWOX in normal mast cell biology, we have generated a conditional model of mast cell-specific WWOX deletion (CPA3-Cre;

WWOXfl/fl Tg mice). We have shown that Wwox expression is significantly decreased in mBMCMCs generated from CPA3-Cre;WWOXfl/fl-Tg mice compared to single transgenic mice nad demonstrate that this functions in a tissue-specific manner. Preliminary data demonstrate that CPA3-Cre;WWOXfl/fl-Tg mice have a decreased lifespan, which is interesting since the complete mast cell deficient CPA3 model does not report decreased lifespan as a phenotypic outcome [83].

2.6 Conclusions & Future Directions

Taken together, this data suggests that Wwox plays a role in mediating the DNA damage response in mast cells, and that loss of WWOX may contribute to radioresistance in malignant canine mast cells. Furthermore, loss of Wwox may also serve as a potential prognostic indicator due to correlative loss observed with increasing histologic grade.

Additional studies to validate the mechanism by which WWOX loss impacts the DDR pathway in the C2 MC line are warranted to better understand these mechanisms and will likely support the known interactions that occur through the ATM-mediated DDR pathway [40]. After processing of the additional higher-grade tumor samples via immunohistochemistry, future studies or retrospective analysis of the prognostic findings of the patient samples expressing decreased Wwox expression may also provide insight 36 into the utilization of WWOX loss in prognostication of primary mast cell tumors.

Additionally, digital image analysis may be utilized to better quantify the TMA results, potentially improving correlation and clinical usability [80, 84, 85].

Due to the challenges of culturing BR cells in the methylcellulose-based media and differences noted in methylcellulose growth in preliminary C2 cell studies, it is possible that manipulation of WWOX in these cell lines influences the cell surface protein expression rather than exerting a major effect on the DDR pathway. WWOX influences cell surface proteins involved in interactions with the tumor microenvironment, specifically integrin-α3, so additional studies examining potential attachment differences in vitro or in vivo may offer more compelling results in understanding additional pathways that WWOX may influence within MCTs [82].

Full phenotyping of the CPA3-Cre;WWOXfl/fl-Tg mouse model is currently being conducted to assess the molecular consequences of tissue specific WWOX ablation on normal mast cell function. We anticipate that full phenotyping may shed light on the decreased lifespan exhibited by the tissue specific WWOX ablation. Through better understanding the role that WWOX plays in both normal and malignant mast cells, we can better understand additional genetic participants in tumorigenesis with the ultimate goal of developing more refined treatment options and prognostic indicators of disease.

These studies have provided a foundation to generate more information surrounding the role of WWOX in both canine neoplasia and normal mast cell function.

Appendix: Tables and Figures

Table 1: Scoring criteria for tumor microarray Code Assigned Criteria for Stain Code Assigned Percent Positive Strength Cells Labeled 0 No Labeling 0 0% labeling 1 Mild cytoplasmic 1 1-25% labeled labeling 2 Moderate 2 26-50% labeled cytoplasmic labeling 3 Strong cytoplasmic 3 51-75% labeled labeling 4 76-100% labeled

Table 2: Primers for quantitative reverse transcriptase polymerase chain reaction

Primers Primer sequence

K9 RT WwoxF Mare 5’-TCCTCCGAGTCCCATAGATTC -3’

K9 RT WwoxR Mare 5’-CGGCAGCAGTTGTTGAAGTA-3’ h18S574F 5’-AAATCCTTTAACGAGGATCCATT-3’ h18S652R 5’-AATATACGCTATTGGAGCTGGA-3’

Table 3: Density of seeding for clonogenic assay

Treatment Dose (Gy) C2 Cells Plated per Well BR Cells Plated per Well

0 125 250

2 250 500

4 500 750

6 1000 1000

Primary Tumors Cell Lines

) 0.004 ✱✱✱ CT

) ✱✱✱✱

ΔΔ 0.002 - CT 2 ΔΔ 0.003 - 2 0.001

0.002 0.00010

0.001 0.00005

0.00000

0.000 Relative Expression ( Relative Expression ( BR C2

cBMCMC cBMCMC Primary Canine MCTs

Figure 1: Wwox transcript is decreased in primary canine MCTs and MC lines qRT-PCR was performed to determine Wwox transcript expression in normal canine

BMCMCs, primary MCTs, and malignant MC lines (BR and C2). All reactions were performed in triplicate and included non-template controls. Relative gene expression was calculated using the comparative threshold cycle method and all samples were normalized to 18S. Data demonstrated that Wwox transcript expression is substantially decreased in primary canine MCTs and malignant MC lines as compared to normal canine BMCMCs. Additionally, this data is demonstrated to be independent of stem cell factor (SCF), which is required for culture of BMCMCs (see supplemental data).

Cell Lines

BR C2 C57 P815

Wwox

β-Actin

Primary Tumor Samples

T1 T2 T3 T4 T5

Figure 2: Wwox protein expression in primary canine MCTs and MC lines Protein lysates were generated from canine MC lines (upper panel) and primary canine MCTs (lower panel). Murine mastocytoma cell lines (C57, P815) served as positive controls for rabbit anti-human Wwox antibody. Representative blots from three experimental replicates are shown.

TMA NOVUS Stain Intensity TMA In House Stain Intensity 3 3 ****

2 2

1 1 Stain Score Stain Score

0 0

Low GradeHigh Grade Low GradeHigh Grade

TMA NOVUS % Cells Stained TMA In House % Cells Stained 4 4 **** **** 3 3

2 2 Percent Score Percent Score 1 1

0 0

Low GradeHigh Grade Low GradeHigh Grade

Figure 3: Loss of Wwox expression is associated with increased histologic tumor grade in primary canine MCTs Representative areas (2 mm) of 31 samples of either normal tissue (n = 15) or mast cell tumor (low-grade n = 14, high-grade n = 5) were selected with pathologist guidance. Immunohistochemical staining for Wwox was performed using normal canine testicular tissue as a positive control. TMA was analyzed by two board-certified veterinary pathologists and scored blindly using the specifications in Table 1. Results

43 were averaged between raters for comparison based on tumor grade. Degree of staining intensity was not found to correlate with histologic grade in the commercially available antibody but did correlate with our in-house antibody such that with high-grade tumors had significantly less stain intensity. The percent of neoplastic cells staining was found to be significantly different for both antibodies such that with high tumor grade, the percent of cells staining for Wwox decreased, which supports our hypothesis that loss of WWOX may lead to a more aggressive tumor phenotype. Weighted Kappa scores were determined using GraphPad Prism online calculator. For the Novus antibody, there was fair interrater agreement for stain intensity (K = 0.393) and moderate agreement for percent of cells stained (K = 0.588). For the in-house antibody, there was fair inerrater agreement for both staining intensity and percent of cells stained (K = 0.264 and 0.273, respectively).

BR WWOX ✱✱✱

)

s 8

e BR Cell Line

1 15

o EV WWOX

x t

e Wwox

d 10

a F

β-Actin

e r

v 5

i o

t Wwox/β-

N 0.11 0.25 a

( Actin ratio

l e

R 0

V X E O R W B W R B

Figure 4: Generation of canine MC lines expressing Wwox lentiviral constructs

Full-length canine WWOX cDNA was PCR amplified from canine testes and cloned into the pCDH-copGFP lentiviral expression plasmid (Systems Biosciences). Canine BR cells were transduced with empty vector (EV) or WWOX lentivirus. FACs-mediated cell sorting based on GFP expression was performed 72 h post-transduction. qRT-PCR and

Western blotting was performed to validate Wwox overexpression in BR cells. Gene expression was calculated using the ΔCt method and all samples were normalized to 18S.

Data represent mean +/- SD of three independent experiments and representative blots from three experimental replicates are shown.

C2 shRNAs ✱

2.0 ✱

)

s 8

e C2 Cell Line

1 1.5

o t

x shWWOX

d 1.0 Scr 554 1304

m Wwox

e r

v 0.5

a (

l β-Actin e

R 0.0 Wwox/β- 0.58 0.21 0.30 e 4 4 Actin ratio l 5 0 b 5 3 m X 1 ra O X c W O s W 2 W C A W N A R N h R s h 2 s C 2 C

Figure 5: Generation of canine MC lines expressing WWOX-targeted shRNA lentiviral constructs

Full-length canine WWOX cDNA was PCR amplified from canine testes and cloned into the pCDH-copGFP lentiviral expression plasmid (Systems Biosciences). Short hairpin

RNAs targeting WWOX were cloned into the pGreenPuro shRNA cloning lentivector

(Systems Biosciences). Canine C2 cells were transduced with scramble control or

WWOX-shRNAs (shWWOX-554 or shWWOX-1304). FACs-mediated cell sorting based on GFP expression was performed 72 h post-transduction. qRT-PCR and Western blotting was performed to validate Wwox knockout in C2 cells. Gene expression was calculated using the ΔCt method and all samples were normalized to 18S. Data represent

47 mean +/- SD of three independent experiments and representative blots from three experimental replicates are shown.

BR MTT

) 15 g

e BR EV

i n

d BR WWOX

e C

S 10

e i

l 5

R N ( 0 0 24 48 72 96 Time (hours)

C2 shRNA MTT

) 8 g

e C2 scramble

i n

d C2 shRNA a

e 6 h

e WWOX 554

C2 shRNA

o 4 WWOX 1304

i a

t 2

R N ( 0 0 24 48 72 96 Time (hours)

Figure 6: Alteration of Wwox does not affect cell viability in BR or C2 mast cells Canine BR cells expressing control vector (EV) or WWOX and C2 cells expressing scramble control or shRNAs targeting WWOX (shWWOX-554, shWWOX-1304) were plated at 5,000 cells/well in 96-well plates and cell viability was assessed at indicated

49 time points using the MTT assay (Roche). All experiments were performed in triplicate and data represent the mean +/- SD of three independent experiments.

BR IR

0.1

BR EV n

o BR WWOX

t c

a 0.01

i v

i 0.001

u S

0.0001 0 2 4 6 Radiation Dose (Gy)

C2 shRNA IR

1 C2 scramble

* n

o C2 shRNA

i t

c WWOX 554

a 0.1 r

F C2 shRNA

g WWOX 1304

i v

i 0.01

u S

0.001 0 2 4 6 Radiation Dose (Gy)

Figure 7: Wwox knockout affects clonogenic survival following ionizing radiation in the canine C2 mast cell line To assess the effects of WWOX manipulation, canine BR cells expressing control vector

(EV) or WWOX or canine C2 cells expressing scramble control or shRNAs targeting

WWOX (shWWOX-554, shWWOX-1304) were assessed for differences clonogenic survival following IR. Cells were analyzed via standard colony formation assay to

51 determine surviving fraction. BR cells did not grow in the methylcellulose-based media, so the impact of WWOX overexpression on clonogenic survival following exposure to IR was unable to be assessed accurately. Therefore, MTT assays were performed on BR cells to assess viability post-IR. No significant differences were noted between EV and

WWOX groups, indicating that restoration of WWOX does not impact cell viability after

IR exposure in the BR cell line. C2 cells expressing shRNAs targeting WWOX were noted to have increased clonogenic survival following radiation at 4 & 6 Gy. This suggests that loss of WWOX confers a clonogenic survival advantage for cells after ionizing radiation exposure at therapeutically relevant levels.

Figure 8: CPA3-Cre;WWOXfl/fl-Tg mouse model validation Transgenic (Tg) mice carrying a conditional WWOX allele (WWOXfl/fl) were crossed with

Cpa3-Cre Tg mice that express Cre recombinase under the carboxypeptidase A3 promoter that restricts expression to mast cells and basophils. qRT-PCR was performed on mBMCMCs generated from CPA3-Cre, CPA3-Cre; WWOXfl/+ or CPA3-Cre;

WWOXfl/fl Tg mice. Protein lysates were generated from lung, spleen, and liver tissues

(upper panel) from CPA3-Cre (wild type, WT) and CPA3-Cre; WWOXfl/fl (knockout, KO) mice. Primary mouse bone marrow-cultured mast cells (mBMCMC) were differentiated in IL-3 (50 ng/mL) for 4-6 weeks from CPA3-Cre, CPA3-Cre; WWOXfl/+ or CPA3-Cre;

WWOXfl/fl Tg mice (lower panel). Protein lysates were separated via SDS-PAGE and western blotting for Wwox and β-actin was performed demonstrating tissue-specific deletion of WWOX. Data represent mean +/- SD of three independent experiments.

Influence of SCF Supplementation

WT +SCF WT +SCF

Wwox

β-Actin

BR C2

Figure 9: Influence of SCF Supplementation BR and C2 cells (~1 x 107) were cultured in normal growth and supplemented with 100 ng/mL canine SCF (Cat #2278SC025, R&D Systems Inc., Minneapolis, MN) for 24 hours at 37ºC supplemented with 5% CO2. Cells were collected and processed to be evaluated for changes in Wwox protein expression. No changes in Wwox protein expression were noted.

References

[1] Bednarek AK, Laflin KJ, Daniel RL, et al. WWOX , a Novel WW Domain-

containing Protein Mapping to Affected in Breast Cancer Advances in Brief.

Cancer Res 2000; 60: 2140–2145.

[2] Durkin SG, Glover TW. Chromosome Fragile Sites. Annu Rev Genet 2007; 41:

169–192.

[3] Del Mare S, Salah Z, Aqeilan RI. WWOX: Its genomics, partners, and functions. J

Cell Biochem 2009; 108: 737–745.

[4] Ludes-Meyers JH, Kil H, Bednarek AK, et al. WWOX binds the specific proline-

rich ligand PPXY: Identification of candidate interacting proteins. Oncogene 2004;

23: 5049–5055.

[5] Salah Z, Aqeilan R, Huebner K. WWOX gene and gene product: tumor

suppression through specific protein interactions. Future oncology (London,

England) 2010; 6: 249–259.

[6] Salah Z, Alian A, Aqeilan RI. WW domain-containing proteins: Retrospectives

and the future. Frontiers in Bioscience 2012; 17: 331–348.

[7] Bray JE, Marsden BD, Oppermann U. The human short-chain

dehydrogenase/reductase (SDR) superfamily: A bioinformatics summary. Chem

Biol Interact 2009; 178: 99–109.

[8] Nunez MI, Ludes-Meyers J, Aldaz CM. WWOX protein expression in normal

human tissues. J Mol Histol 2006; 37: 115–125.

[9] Breitbach J, Fenger JM. Unpublished TMA Data.

[10] Aqeilan RI, Trapasso F, Hussain S, et al. Targeted deletion of Wwox reveals a

tumor suppressor function. Proc Natl Acad Sci 2007; 104: 3949–3954.

[11] Aqeilan RI, Hagan JP, De Bruin A, et al. Targeted ablation of the WW domain-

containing oxidoreductase tumor suppressor leads to impaired steroidogenesis.

Endocrinology 2009; 150: 1530–1535.

[12] Aqeilan RI, Hassan MQ, De Bruin A, et al. The WWOX tumor suppressor is

essential for postnatal survival and normal bone metabolism. J Biol Chem 2008;

283: 21629–21639.

[13] Abdeen SK, Del Mare S, Hussain S, et al. Conditional inactivation of the mouse

Wwox tumor suppressor gene recapitulates the null phenotype. J Cell Physiol

2013; 228: 1377–1382.

[14] Abu-Remaileh M, Aqeilan RI. Tumor suppressor WWOX regulates glucose

metabolism via HIF1α modulation. Cell Death Differ 2014; 21: 1805–1814.

[15] Dayan S, O’Keefe L V., Choo A, et al. Common chromosomal fragile site

FRA16D tumor suppressor WWOX gene expression and metabolic reprograming

in cells. Genes Chromosom Cancer 2013; 52: 823–831.

[16] O’Keefe L V, Colella A, Dayan S, et al. Drosophila orthologue of WWOX, the

chromosomal fragile site FRA16D tumour suppressor gene, functions in aerobic

metabolism and regulates reactive oxygen species. Hum Mol Genet 2011; 20: 497–

509.

[17] Mallaret M, Synofzik M, Lee J, et al. The tumour suppressor gene WWOX is

mutated in autosomal recessive cerebellar ataxia with epilepsy and mental

retardation. A J Neurol. DOI: 10.1093/brain/awt338.

[18] Piard J, Hawkes L, Milh M, et al. The phenotypic spectrum of WWOX -related

disorders: 20 additional cases of WOREE syndrome and review of the literature.

Genet Med 2019; 21: 1308–1318.

[19] Tanna M, Aqeilan RI. Modeling WWOX loss of function in vivo: What have we

learned? Frontiers in Oncology 2018; 8: 420.

[20] Krummel KA, Denison SR, Calhoun E, et al. The common fragile site FRA16D

and its associated gene WWOX are highly conserved in the mouse at Fra8E1.

Genes Chromosom Cancer 2002; 34: 154–167.

[21] Arlt MF, Casper AM, Glover TW. Common fragile sites. Cytogenet Genome Res

2003; 100: 92–100.

[22] Sarni D, Kerem B. The complex nature of fragile site plasticity and its importance

in cancer. Current Opinion in Cell Biology 2016; 40: 131–136.

[23] Pandis N, Heim S, Bardi G, et al. Whole‐arm t(1;16) and i(1q) as sole anomalies

identify gain of 1 q as a primary chromosomal abnormality in breast cancer.

Genes, Chromosom Cancer 1992; 5: 235–238.

[24] Chen T, Sahin A, Aldaz CM. Deletion map of chromosome 16q in ductal

carcinoma in situ of the breast: Refining a putative tumor suppressor gene region.

Cancer Res 1996; 56: 5605–5609.

[25] Latil A, Cussenot O, Fournier G, et al. Loss of heterozygosity at chromosome 16q

in prostate adenocarcinoma: identification of three independent regions. Cancer

Res 1997; 57: 1058–62.

[26] Abdeen SK, Salah Z, Maly B, et al. Wwox inactivation enhances mammary

tumorigenesis. Oncogene 2011; 30: 3900–3906.

[27] Salah Z, Aqeilan R, Huebner K. WWOX gene and gene product: tumor

suppression through specific protein interactions. Future oncology (London,

England) 2010; 6: 249–259.

[28] Aqeilan RI, Croce CM. WWOX in Biological Control and Tumorigenesis. J Cell

Physiol 2007; 731: 720–731.

[29] Yan H, Tong J, Lin X, et al. Effect of the WWOX gene on the regulation of the

cell cycle and apoptosis in human ovarian cancer stem cells. Mol Med Rep 2015;

12: 1783–1788.

[30] Kuroki T, Yendamuri S, Trapasso F, et al. The tumor suppressor gene WWOX at

FRA16D is involved in pancreatic carcinogenesis. Clin Cancer Res 2004; 10:

2459–65.

[31] Iliopoulos D, Guler G, Han SY, et al. Fragile genes as biomarkers: Epigenetic

control of WWOX and FHIT in lung, breast and bladder cancer. Oncogene 2005;

24: 1625–1633.

[32] Wang X, Chao L, Jin G, et al. Association between CpG island methylation of the

WWOX gene and its expression in breast cancers. Tumor Biol 2009; 30: 8–14.

[33] Pospiech K, Płuciennik E, Bednarek AK. WWOX Tumor Suppressor Gene in

Breast Cancer, a Historical Perspective and Future Directions. Front Oncol 2018;

8: 1–12.

[34] Bednarek AK, Keck-Waggoner CL, Daniel RL, et al. WWOX, the FRA16D gene,

behaves as a suppressor of tumor growth. Cancer Res 2001; 61: 8068–8073.

[35] Aqeilan RI, Pekarsky Y, Herrero JJ, et al. Functional association between Wwox

tumor suppressor protein and p73, a p53 homolog. Proc Natl Acad Sci 2004; 101:

4401–4406.

[36] Aqeilan RI, Donati V, Palamarchuk A, et al. WW Domain–Containing Proteins,

WWOX and YAP, Compete for Interaction with ErbB-4 and Modulate Its

Transcriptional Function. Cancer Res 2005; 65: 6764–6772.

[37] Yang J, Cogdell D, Yang D, et al. Deletion of the WWOX gene and frequent loss

of its protein expression in human osteosarcoma. Cancer Lett 2010; 291: 31–38.

[38] Kurek KC, Del Mare S, Salah Z, et al. Frequent attenuation of the WWOX tumor

suppressor in osteosarcoma is associated with increased tumorigenicity and

aberrant RUNX2 expression. Cancer Res 2010; 70: 5577–5586.

[39] Del Mare S, Aqeilan RI. Tumor Suppressor WWOX inhibits osteosarcoma

metastasis by modulating RUNX2 function. Sci Rep 2015; 5: 1–9.

[40] Abu-Odeh M, Salah Z, Herbel C, et al. WWOX, the common fragile site FRA16D

gene product, regulates ATM activation and the DNA damage response. Proc Natl

Acad Sci 2014; 111: E4716–E4725.

[41] Burma S, Chen BP, Murphy M, et al. ATM Phosphorylates Histone H2AX in

Response to DNA Double-strand Breaks. J Biol Chem 2001; 276: 42462–42467.

[42] Sancar A, Lindsey-Boltz LA, Ünsal-Kaçmaz K, et al. Molecular Mechanisms of

Mammalian DNA Repair and the DNA Damage Checkpoints. Annu Rev Biochem

2004; 73: 39–85.

[43] Abu-Odeh M, Hereema NA, Aqeilan RI. WWOX modulates the ATR-mediated

DNA damage checkpoint response. Oncotarget; 7. Epub ahead of print 2015. DOI:

10.18632/oncotarget.6571.

[44] Priester WA. Skin tumors in domestic animals. Data from 12 United States and

Canadian colleges of veterinary medicine. J Natl Cancer Inst 1973; 50: 457–66.

[45] Cohen D, Reif JS, Brodey RS, et al. Epidemiological Analysis of the Most

Prevalent Sites and Types of Canine Neoplasia Observed in a Veterinary Hospital.

Cancer Res 1974; 34: 2859–2868.

[46] Villamil JA, Henry CJ, Bryan JN, et al. Identification of the most common

cutaneous neoplasms in dogs and evaluation of breed and age distributions for

selected neoplasms. J Am Vet Med Assoc 2011; 239: 960–965.

[47] Shoop SJ, Marlow S, Church DB, et al. Prevalence and risk factors for mast cell

tumours in dogs in England. Canine Genet Epidemiol 2015; 2: 1.

[48] Kiupel M, Webster JD, Bailey KL, et al. Proposal of a 2-tier histologic grading

system for canine cutaneous mast cell tumors to more accurately predict biological

behavior. Vet Pathol 2011; 48: 147–155.

[49] O’Keefe DA. Canine mast cell tumors. The Veterinary clinics of North America.

Small animal practice 1990; 20: 1105–1115.

[50] Patnaik AK, Ehler WJ, MacEwen EG. Canine Cutaneous Mast Cell Tumor:

Morphologic Grading and Survival Time in 83 Dogs. Vet Pathol 1984; 21: 469–

474.

[51] Northrup NC, Harmon BG, Gieger TL, et al. Comparative Oncol-ogy Program,

College of Veterinary Medicine, University of Geor-gia. Cheng,

https://journals.sagepub.com/doi/pdf/10.1177/104063870501700305 (2005,

accessed 3 July 2019).

[52] Stefanello D, Buracco P, Sabattini S, et al. Comparison of 2- and 3-category

histologic grading systems for predicting the presence of metastasis at the time of

initial evaluation in dogs with cutaneous mast cell tumors: 386 cases (2009–2014).

J Am Vet Med Assoc 2015; 246: 765–769.

[53] Scase TJ, Edwards D, Miller J, et al. Canine mast cell tumors: Correlation of

apoptosis and proliferation markers with prognosis. J Vet Intern Med 2006; 20:

151–158.

[54] Şmiech A, Şlaska B, Łopuszyński W, et al. Epidemiological assessment of the risk

of canine mast cell tumours based on the Kiupel two-grade malignancy

classification. Acta Vet Scand 2018; 60: 70.

[55] Warland J, Dobson J. Breed predispositions in canine mast cell tumour: A single

centre experience in the United Kingdom. Vet J 2013; 197: 496–498.

[56] Webster JD, Yuzbasiyan-Gurkan V, Kaneene JB, et al. The Role of c-KIT in

Tumorigenesis: Evaluation in Canine Cutaneous Mast Cell Tumors. Neoplasia

2006; 8: 104–111.

[57] Letard S, Yang Y, Hanssens K, et al. Gain-of-Function Mutations in the

Extracellular Domain of KIT Are Common in Canine Mast Cell Tumors. Mol

Cancer Res 2008; 6: 1137–1145.

[58] London CA, Malpas PB, Wood-Follis SL, et al. Multi-center, placebo-controlled,

double-blind, randomized study of oral toceranib phosphate (SU11654), a receptor

tyrosine kinase inhibitor, for the treatment of dogs with recurrent (either local or

distant) mast cell tumor following surgical excision. Clin Cancer Res 2009; 15:

3856–3865.

[59] Hahn KA, Legendre AM, Shaw NG, et al. Evaluation of 12-and 24-month survival

rates after treatment with masitinib in dogs with nonresectable mast cell tumors

AbbreviAtions CI Confidence interval IC 50 Median inhibitory concentration MCT

Mast cell tumor TKI Tyrosine kinase inhibitor TTP Time t,

https://avmajournals.avma.org/doi/pdf/10.2460/ajvr.71.11.1354 (2010, accessed 16

July 2019).

[60] Blacklock KB, Birand Z, Biasoli D, et al. Identification of molecular genetic

contributants to canine cutaneous mast cell tumour metastasis by global gene

expression analysis. PLoS One; 13. Epub ahead of print 2018. DOI:

10.1371/journal.pone.0208026.

[61] Giantin M, Granato A, Baratto C, et al. Global gene expression analysis of Canine

cutaneous mast cell tumor: Could molecular profiling be useful for subtype

classification and prognostication? PLoS One; 9. Epub ahead of print 2014. DOI:

10.1371/journal.pone.0095481.

[62] Mochizuki H, Thomas R, Moroff S, et al. Genomic profiling of canine mast cell

tumors identifies DNA copy number aberrations associated with KIT mutations

and high histological grade. Chromosom Res 2017; 25: 129–143.

[63] Dorn CR, N Taylor DO, Schneider R, et al. Survey of Animal Neoplasms In

Alameda and Contra Costa Counties, California. II. Cancer Morbidity in DOls

and Cats From Alameda Countyl,2, https://academic.oup.com/jnci/article-

abstract/40/2/307/929183 (accessed 23 April 2020).

[64] Downing S, Chien MB, Kass PH, et al. Prevalence and importance of internal

tandem duplications in exons 11 and 12 of c-kit in mast cell tumors of dogs. Am J

Vet Res 2002; 63: 1718–1723.

[65] Zemke D, Yamini B, Yuzbasiyan-Gurkan V. Mutations in the Juxtamembrane

Domain of c-KIT Are Associated with Higher Grade Mast Cell Tumors in Dogs.

Vet Pathol 2002; 39: 529–535.

[66] Pulz LH, Barra CN, Alexandre PA, et al. Identification of two molecular subtypes

in canine mast cell tumours through gene expression profiling. PLoS One; 14.

Epub ahead of print 1 June 2019. DOI: 10.1371/journal.pone.0217343.

[67] Aqeilan RI, Palamarchuk A, Weigel RJ, et al. Physical and functional interactions

between the Wwox tumor suppressor protein and the AP-2γ transcription factor.

Cancer Res 2004; 64: 8256–8261.

[68] Chang R, Song L, Xu Y, et al. Loss of Wwox drives metastasis in triple-negative

breast cancer by JAK2/STAT3 axis. Nat Commun 2018; 9: 1–12.

[69] Fenger JM, Bear MD, Volinia S, et al. Overexpression of miR-9 in mast cells is

associated with invasive behavior and spontaneous metastasis. BMC Cancer 2014;

14: 84.

[70] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-

time quantitative PCR and the 2-ΔΔCT method. Methods 2001; 25: 402–408.

[71] Franken NAP, Rodermond HM, Stap J, et al. Clonogenic assay of cells in vitro.

Nat Protoc 2006; 1: 2315–2319.

[72] Ludes-Meyers JH, Kil H, Parker-Thornburg J, et al. Generation and

characterization of mice carrying a conditional allele of the Wwox tumor

suppressor gene. PLoS One 2009; 4: 7775.

[73] Lilla JN, Chen CG, Mukai K, et al. Reduced mast cell and basophil numbers and

function in Cpa3-Cre; Mcl-1 fl/fl mice. Blood 2011; 118: 6930–6938.

[74] Landis JR, Koch GG. The Measurement of Observer Agreement for Categorical

Data. Biometrics 1977; 33: 159.

[75] García G, Brazís P, Majó N, et al. Comparative morphofunctional study of

dispersed mature canine cutaneous mast cells and BR cells, a poorly differentiated

mast cell line from a dog subcutaneous mastocytoma. Vet Immunol Immunopathol

1998; 62: 323–37.

[76] Devinney R, Gold WM. Establishment of Two Dog Mastocytoma Cell Lines in

Continuous Culture. 1990.

[77] Ding-E Young J, Liu C-C, Butler G, et al. Identification, purification, and

characterization of a mast cell-associated cytolytic factor related to tumor

necrosis factor (cell-mediated immunity/natural cytotoxic cell/natural killer

cell/cytotoxicity/pore-forming protein). 1987.

[78] Jonmarker Jaraj S, Camparo P, Boyle H, et al. Intra- and interobserver

reproducibility of interpretation of immunohistochemical stains of prostate cancer.

Virchows Arch 2009; 455: 375–381.

[79] Borlot VF, Biasoli I, Schaffel R, et al. Evaluation of intra- and interobserver

agreement and its clinical significance for scoring bcl-2 immunohistochemical

expression in diffuse large B-cell lymphoma. Pathol Int 2008; 58: 596–600.

[80] Rizzardi AE, Johnson AT, Vogel RI, et al. Quantitative comparison of

immunohistochemical staining measured by digital image analysis versus

pathologist visual scoring. Diagn Pathol 2012; 7: 42.

[81] Garrett L. Canine mast cell tumors: diagnosis, treatment, and prognosis. Vet Med

Res Reports 2014; 5: 49.

[82] Gourley C, Paige AJW, Taylor KJ, et al. WWOX gene expression abolishes

ovarian cancer tumorigenicity in vivo and decreases attachment to fibronectin via

integrin α3. Cancer Res 2009; 69: 4835–4842.

[83] Feyerabend TB, Weiser A, Tietz A, et al. Cre-mediated cell ablation contests mast

cell contribution in models of antibody- and T cell-mediated autoimmunity.

Immunity 2011; 35: 832–844.

[84] Rizzardi AE, Zhang X, Vogel RI, et al. Quantitative comparison and

reproducibility of pathologist scoring and digital image analysis of estrogen

receptor β2 immunohistochemistry in prostate cancer. Diagn Pathol 2016; 11: 63.

[85] Gavrielides MA, Gallas BD, Lenz P, et al. Observer variability in the interpretation

of HER2/neu immunohistochemical expression with unaided and computer-aided

digital microscopy. Arch Pathol Lab Med 2011; 135: 233–242.