Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma

ARTICLE IN PRESS The Veterinary Journal ■■ (2014) ■■–■■

Contents lists available at ScienceDirect

The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l

Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma Lorella Maniscalco *, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria Department of Veterinary Sciences, University of Turin, Largo Braccini 2, Grugliasco, TO, Italy

A R T I C L E

I N F O

Article history: Accepted 5 September 2014 Keywords: Dog Osteosarcoma Comparative oncology Insulin-like growth factor 1 receptor (IGF1R) Therapeutic target Targeted therapy

A B S T R A C T

Insulin-like growth factor 1 receptor (IGF-1R) is a cell membrane receptor widely expressed in tissues and involved in different cancers in humans. IGF-1R expression in human osteosarcoma has been associated with the development of tumour metastasis and with prognosis, and represents an attractive therapeutic target. The goal of this study was to investigate the expression of IGF-1R in canine osteosarcoma tissues and cell lines and assess its role and prognostic value. Samples from 34 dogs were examined by immunohistochemistry for IGF-1R expression. IGF-1R/AKT/MAPK signalling was evaluated by western blot and quantitative polymerase chain reaction in the cell lines. In addition, the in vitro inhibition of IGF-1R with pycropodophillin (PPP) was used to evaluate molecular and biological effects. Immunohistochemical data showed that IGF-1R was expressed in 71% of the analysed osteosarcoma samples and that dogs with higher levels of IGF-IR expression (47% of cases) had decreased survival (P < 0.05) when compared to dogs with lower IGF-IR expression. Molecular studies demonstrated that in canine osteosarcoma IGF-IR is activated by IGF-1 mostly in a paracrine or endocrine (rather than autocrine) manner, leading to activation of AKT/MAPK signalling. PPP caused p-IGF-1R dephosphorylation with partial blocking of p-MAPK and p-AKT, as well as apoptosis. It was concluded that IGF-1R is expressed and plays a role in canine osteosarcoma and that its expression is correlated with a poor prognosis. As in humans, IGF-1R may represent a good therapeutic target and a prognostic factor for canine osteosarcoma. © 2014 Elsevier Ltd. All rights reserved.

Introduction Canine and feline spontaneous tumours are suitable models for human cancers due to strong molecular, histopathological and clinical similarities. As such, dogs and cats represent useful alternative models to rodents for human cancer biology and translational therapy (Withrow et al., 1991; Morello et al., 2011). Osteosarcoma (OSA), the most common primary malignant bone tumour in dogs, is locally aggressive, and has a high metastatic potential and a poor prognosis (Farese et al., 2009; Morello et al., 2011; Casteleyn et al., 2013). Several studies have demonstrated similarities between canine and human OSA at the molecular level by showing comparable expression of different proteins. Canine OSA cell lines and tumours frequently contain mutations that inactivate p53 (Johnson et al., 1998) and the phosphatase and tensin homologue (PTEN) family of tumour suppressor genes (Levine et al., 2002). Insulin and

* Corresponding author. Tel.: +39 011 670 9037. E-mail address: [email protected] (L. Maniscalco).

hepatocyte growth factors influence tumour growth, invasion and malignant phenotype in canine OSA cell lines (MacEwen et al., 2003, 2004). Matrix metalloproteinase 2 (MMP-2) expression has been detected only in high-grade OSAs, while pro-MMP-9 production correlated with the histological grade of OSA, suggesting a potential role of MMPs in the pathogenesis of canine OSA growth and metastasis (Loukopoulos et al., 2003). STAT3 activation has also been shown to contribute to the survival and proliferation of human and canine OSA cell lines (Lana et al., 2000; Loukopoulos and Robinson, 2007; Fossey et al., 2009). Recently, the role and expression of the tyrosine kinase receptors (TKRs) have been investigated in veterinary oncology. Overexpression of the erb-B2 gene, encoding for human epidermal growth factor receptor 2 (HER2), was observed in 86% and 40% of canine OSA cell lines and tissue samples, respectively (Flint et al., 2004), suggesting an involvement in the pathogenesis of this tumour like in humans (Rainusso et al., 2012). Furthermore, as in humans, 79% of canine OSA samples overexpress the MET oncogenes; cell motility and invasiveness appear to be MET-dependent since they can be inhibited by small interfering RNAs (siRNAs) that are specific for MET (De Maria et al., 2009). Finally, we have recently reported

http://dx.doi.org/10.1016/j.tvjl.2014.09.005 1090-0233/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

2

that PDGF receptors are overexpressed in canine OSA, adding a new potential therapeutic target (Maniscalco et al., 2013). Insulin-like growth factor-1 receptor (IGF-1R) is a transmembrane TKR consisting of two extracellular α-subunits responsible for ligand binding, two β-subunits with a transmembrane domain and an intracellular tyrosine kinase COOH-terminal domain (Werner et al., 2008). The specific interaction between IGF-1 and IGF-1R induces the phosphorylation of intracellular tyrosine residue (βsubunit) and results in activation of the receptor. The activated form of IGF-1R is able to activate the PI3K/AKT and mitogen-activated kinase signalling (MAPK) pathways (Girnita et al., 2004). In humans, the IGF pathway is involved in bone homeostasis, since it regulates the proliferation of osteoblasts (Giustina et al., 2008; Zhang et al., 2012; Kubota et al., 2013), but it is also able to promote tumour initiation and progression (Baserga, 1995; Avnet et al., 2009) and it is implicated in radio- and chemoresistance (Bohula et al., 2003). In human OSA, IGF-1R is correlated with tumour metastasis and a poor prognosis (Wang et al., 2012) and in vitro inhibition of IGF1R by specific tyrosine kinase inhibitors (TKIs) induces apoptosis in multidrug resistant OSA cell lines (Duan et al., 2009). Based on these data, specific inhibitors of the IGF-1R pathway have been investigated and the cyclolignan picropodophyllin (PPP) molecule, a potent and selective IGF-1R inhibitor, is currently undergoing clinical trials in humans (Wu et al., 2013). The advantages of this selective compound are that it does not react with insulin receptors and induces very limited tumour resistance (Vasilcanu et al., 2006). In veterinary medicine, IGF-1R has been studied in canine mammary tumours (Klopfleisch et al., 2010; Dolka et al., 2011), in melanomas (Thamm et al., 2010) and in insulinomas (Buishand et al., 2012) as well as in OSA cell lines (MacEwen et al., 2004), However, there are no data about its role in canine OSA as a prognostic factor and as a potential specific target for TKIs. The aim of this study was to investigate IGF-1R expression in canine OSA tissues and cell lines, to evaluate its prognostic value in relation to clinical outcome, and to test the in vitro effects of IGF-1R inhibition in canine OSA cell lines.

Total RNA was obtained from all OSA cell lines and cDNA was synthesised from 1 μg of total RNA using the QuantiTect Reverse Transcription kit (Qiagen) including genomic DNA removal reagent. To determine the relative amounts of specific IGF-1 and IGF-1R transcripts, the cDNA was subjected to quantitative PCR using the IQ SYBR Green Supermix (BioRad) and the IQ5 detection system (BioRad). Forward and reverse primers employed were: IGF-1: 5′-CCAAGGCTCAGAAGGAAGTACA-3′, 5′-CTTCACAACTCAGGAAGGTCT-3′; IGF-1R: 5′-CCGACGAGTGGAGAAATCTGT-3′, 5′GGAGGTAGCCCTCGATCACT-3′; GAPDH: 5′-GGCACAGTCAAGGCTGAGAAC-3′, 5′CCAGCATCACCCCATTTGAT-3′. The level of gene expression was calculated using a relative quantification assay corresponding to the comparative cycle threshold (Ct) method: the amount of target, normalised to the endogenous housekeeping gene (GAPDH) and relative to the calibrator (control sample), was then transformed by 2ΔΔCt (fold increase), where ΔΔCt = ΔCt (sample) − ΔCt (control) and ΔCt is the Ct of the target gene subtracted from the Ct of the housekeeping gene. GAPDH was selected as housekeeping gene among the six following common endogenous normalisers genes (HPRT, GAPDH, RP13a, β-actin, RP18s, RP32: Appendix: Supplementary Material S2).

Materials and methods

Western blot analysis

Study population

Total proteins were obtained from primary canine OSA cell lines, OSB1 (De Maria et al., 2009; Maniscalco et al., 2013) and NMuUG cell lines by boiling lysis buffer containing 1% SDS and 0.1 M Tris–HCl (pH 6.8). Total proteins from each sample (40 μg) were separated on an 8% sodium dodecyl sulphate (SDS)–polyacrylamide (PAGE) gel, transferred onto a Hybond-C Extra membrane (Amersham Biosciences) and incubated overnight at 4 °C with the primary antibodies (Table 2). After incubation with a horseradish peroxidase (HRP)-linked secondary antibody diluted 1:2000 in PBS– Tween, membranes were washed in PBS–Tween for 30 min and incubated with an enhanced chemiluminescence reagent (Super Signal West Pico Mouse IgG Detection Kit, Thermo Scientific). Murine NMuUG cell line was used as positive control for IGF-1R and p-IGF-1R expression.

Thirty-four dogs diagnosed with canine appendicular OSA and treated in the period 2007–2011 at the Department of Veterinary Sciences (University of Turin) were included in the study (Table 1). Data for the dogs included history, physical examination, complete blood and urine work-up, and abdominal ultrasound. In addition, limb and chest (two and three views, respectively) radiographic evaluation to examine features and extent of the tumour and to exclude lung metastases was performed; computed tomography (CT) was additionally available and used during the latter 1.5 years of the study. Dogs with pulmonary or lymph node metastases were excluded and in case of regional lymph node enlargement, aspiration and cytology were performed to exclude metastases. All dogs included were treated surgically (amputation or limbsparing) before receiving adjuvant chemotherapy using doxorubicin (30 mg/m2, four to five administrations, 21 days apart) or cisplatin (70 mg/m2, four to five administrations, 21 days apart) as single agents or in combination (4 cycles, 21 days apart, each cycle consisting of cisplatin 50 mg/m2 at day 1 and doxorubicin 15 mg/m2 at day 2) (Appendix: Supplementary Material S1). Follow-up monitoring consisted of clinical and radiographic checks every 3 months during the first year, then every 6 months for a minimum of 2 years. Tissue samples collected postoperatively were fixed in 10% buffered formalin for at least 24 h, embedded in paraffin, sectioned (4 μm thick) and stained with haematoxylin/eosin (H&E). Histological diagnosis was based on WHO guidelines (Slayter et al., 1994) and the grade on Loukopoulos and Robinson’s (2007) system. Immunohistochemical analysis In all cases immunohistochemical (IHC) analysis was carried out on 4-μmthick paraffin sections. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in methanol for 30 min at room temperature. Sections were subjected to high temperature antigen unmasking by incubation for 30 min with citric acid buffer

(pH 6) at 98 °C and all sections were processed in an automated immunostainer (Dako) using an immunohistochemical detection substrate (Dako REAL EnVision Detection System, Peroxidase/DAB+, Rabbit/Mouse; Dako). A rabbit polyclonal antihuman IGF-1R antibody was used (Table 2). Canine normal kidney and adipose tissue were used as positive and negative controls, respectively. A ‘no primary antibody control’ on canine kidney was also employed. Specificity of the primary antibody was evaluated by incubating the primary antibody with excess of the specific peptide used to immunise rabbits. Immunolabelled slides were randomised and masked for blind examination, which was performed independently by two observers (LM, SI). In case of disagreement (2/34 cases; 6%), a consensus was reached using a multi-head microscope. Cytoplasmic immunolabelling was evaluated in neoplastic cells with a semi-quantitative method (Wang et al., 2012).

Cell lines Seven primary canine OSA cell lines (Wall, Penny, Dark, Sky, Pedro, Lord and Desmon) (Maniscalco et al., 2013) and normal osteoblasts (OSB1) were cultured in Iscove’s standard medium supplemented with 10% fetal bovine serum (FBS), 1% glutamine, 100 μg/mL penicillin and 100 μg/mL streptomycin. NMuMG cells were purchased from American Type Culture Collection (ATCC) and cultured in DMEM medium containing 10% FBS (HyClone), 10 μg/mL insulin (Sigma-Aldrich), and 100 IU/ mL penicillin.

Total RNA extraction and quantitative PCR expression analysis

IGF-1R inhibition in vitro The ‘Penny’ and ‘Wall’ cell lines were treated with PPP dissolved in DMSO (Calbiochem, cod. 407247) at 0.001 μM, 0.01 μM, 0.1 μM, 0.25 μM or 0.5 μM for 24 and 48 h in complete medium. Cytotoxicity was evaluated using the MTT method (Tada et al., 1986). All experiments were performed in triplicate. OSA cells (2 × 103) were seeded in 96-well plates and incubated with media (negative control), DMSO (vehicle control) and increasing concentrations of PPP for both 24 and 48 h. Wells with media only were included as controls. Levels of caspase-3/7 activity were determined using the SensoLyte Homogeneous AMC Caspase-3/7 Assay kit (Anaspec) according to the manufacturer’s instructions in a 96-well plate. In the caspase assay, a positive control consisting of 1 × 105 Jurkat cells/well treated with camptothecin (0.001, 0.01, 0.1 and 1 μM) for 5 h (Appendix: Supplementary Material S3) was used and the negative control was the untreated Penny cell line. The vehicle control consisted of the Penny cell line with DMSO alone seeded at 3 × 103 cells/well. After incubation, caspase 3/7 substrate solution was added to all wells and incubated prior to measuring fluorescence on Victor 31420 Multilabel Counter (PerkinElmer) using excitation at 354 nm and emission detection at 442 nm. Levels of caspase 3/7 activity were reported after subtraction of fluorescence levels of wells with media only.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

3

Table 1 Clinicopathological data and immunohistochemical of the dogs included in the study. Case number

Breed

Age (years)

Weight

Sex

Diagnosis

Localisation

Grade

ALP (UI/L)

IV invasion

DFI (days)

OS (days)

160 PM 396 LR 180* 167 PM 272 PM 309 PM 623 PM 369 PM 503* 214 LN 306 PM 1033* 496*

178

High

410

High

180* 211

High Low

297

High

335

Negative

715

High

391

Negative

503* 1089

Low Negative

1

Boxer

2

35

M

Osteoblastic productive

Metatarsus

2

65

No

2

Boxer

4

34

F

Giant cell type

Distal ulna

1

125

No

3 4

Great dane Rottweiler

8 2.5

48 52

M M

Osteoblastic productive Osteoblastic unproductive

Distal radius Distal radius

1 2

58 200

No No

5

Boxer

8

32

F

Osteoblastic productive

Scapula

1

78

Yes

6

7

37

F

Osteoblastic Productive

Distal femur

3

173

Yes

7

German shepherd Rottweiler

7

54

F

Osteoblastic productive

Proximal femur

2

61

No

8

Rottweiler

7.5

52

F

Osteoblastic productive

Distal radius

1

78

No

9 10

Mixed Mixed

8 4

38 43

F M

Osteoblastic unproductive Osteoblastic productive

Proximal humerus Distal radius

2 1

89 48

No Yes

11

Mixed

10

36

M

Chondroblastic

Distal tibia

1

55

No

12 13

11 8

42 38

M M

Fibroblastic Fibroblastic

Proximal humerus Proximal humerus

1 2

84 37

No No

14

Newfoundland German shepherd Mixed

8

52

M

Osteoblastic productive

Distal radius

2

145

No

15

Boxer

8

36

F

Chondroblastic

Distal radius

3

36

No

16

Doberman

8

33

M

Chondroblastic

Distal femur

1

165

No

17

Mixed

10

39

F

Mixed

Distal radius

2

136

Yes

18

Newfoundland

6

53

F

Chondroblastic

Distal femur

2

88

Yes

19

Saint bernard

6

55

M

Osteoblastic productive

Distal radius

1

46

Yes

20 21

Siberian husky Malinois

12 6

38 31

F M

Osteoblastic productive Chondroblastic

Proximal humerus Distal tibia

3 1

210 99

Yes No

22 23

Great dane Great dane

5 4

65 61

M M

Osteoblastic unproductive Fibroblastic

Distal tibia Distal radius

2 2

45 101

No No

24

Mixed

7

40

M

Giant cell

Distal ulna

2

30

No

25

Maremma shepherd Mixed

9

51

M

Osteoblastic productive

Distal radius

2

84

No

7.5

49

F

Osteoblastic productive

Tibia

3

22

No

7

70

M

Osteoblastic productive

Distal radius

3

156

No

28 29

New found retriever Saint Bernard Boxer

7 8

73 38

M M

Osteoblastic productive Osteoblastic productive

Distal radius Distal femur

2 2

55 82

No No

30

Beauceron

6

46

M

Osteoblastic productive

Radius

3

27

No

31

Boxer

8

37

F

Chondroblastic

Radius

3

190

No

32

Boxer

7

35

M

Osteoblastic productive

Femur

3

120

No

33

Greyhound

6

52

Mn

Osteoblastic productive

Radius

3

250

No

34

Labrador

7

40

M

Osteoblastic productive

Radius

3

46

No

26 27

185 PM 103 PM 115 PM 297 PM 436 PM 251 PM 123* 438 PM 1250* 89 PM 553 PM 261 LN 242 PM 126 PM 203* 267 PM 90 PM 60 LR 90 PM 50 PM 176 PM

IGF-1R expression

335

High

1033* 496*

Low Low

235

Low

169

High

176

High

310

High

573

Negative

278

Negative

123* 449

High High

1250* 168

Negative Negative

578

Negative

270

Low

252

Low

200

High

203* 295

Negative Negative

150

High

120

Low

100

High

60

High

245

High

ALP, serum alkaline phosphatase; PM, pulmonary metastases; LR, local recurrence; LN, lymph node metastases. * Cause not related to OSA (censored data).

Western blot analysis of total IGF-1R, p-AKT and p-MAPK was performed on protein lysate obtained from Penny cells treated with PPP at 0.01 μM, 0.1 μM, 0.25 μM or 0.5 μM for 24 and 48 h. In addition, IGF-1R, p-IGF-1R, p-AKT, p-MAPK were evaluated after treatment with and without IGF-1 (1 μM) and PPP (0.5 μM) after 2 h of treatment in Penny and Sky cell lines cultured in medium supplemented with bovine serum albumin (BSA, 0.4%),.

Statistical analysis IHC results and clinicopathological findings were grouped into contingency tables and analysed using Fisher’s exact test or Chi2 test. Survival curves were computed using the Kaplan–Meier method and tests for differences in survival considering all known prognostic factors for canine OSA were performed using the log rank test.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

4

Table 2 Antibodies used in the study. Antibody

Type

Code

Source

IHC

IGF-1 IGF-1R p-IGF1R AKT p-AKT p-MAPK α-Tubulin

Mouse monoclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal Rabbit polyclonal Mouse monoclonal

Sc-9013 Sc-712 3021S 9272 3787 M-7683 T-5168

Santa Cruz Biotechnology Santa Cruz Biotechnology Cell Signaling Technology Cell Signaling Technology Cell Signaling Technology Sigma Aldrich Sigma Aldrich

/ 1:200 / / / / /

WB 1:1,000 1:1,000 1:1,000 1:1,000 1:1,000 1:1,000 1:10,000

IHC, immunohistochemistry; WB, western blot.

All parameters with P < 0.15 on univariate analysis (log rank test) were subjected to multivariate analysis using Cox proportional hazard regression model. Overall survival (OS) was the number of days between surgery and death and the diseasefree interval (DFI) was the number of days between surgery and tumour recurrence and/or evidence of metastasis. Censored data were considered those cases that were still alive or that did not presented tumour recurrence at the end of the monitoring period or that died for unrelated causes. Caspase 3/7 activity and q-PCR data were analysed using Student’s t test. Data were analyzed with MedCalc Statistical Software v.13.3 (MedCalc Software bvba) and P < 0.05 was considered statistically significant.

Results Epidemiological and clinical data Data were collected from 34 cases of appendicular OSA surgically removed from dogs with a mean age of 7 ± 2 years (range, 2–12 years), predominantly males (59%). Seven dogs (21%) were of mixed breed and 27 (79%) were of pure breed. The most represented breeds were Boxers (seven dogs) and Rottweilers (three dogs). The mean DFI was 298 ± 258 days (median, 228), and the mean OS was 264 ± 285 days (median, 274). Clinicopathological findings are listed in Table 1 and details regarding surgical therapy and chemotherapy for each subject are listed in Supplementary Material S1 (Appendix). Histopathology Of the 34 cases, 19 (56%) were osteoblastic productive OSA, 7 (21%) were chondroblastic OSA, 3 (9%) were fibroblastic OSA, 3 (9%) were unproductive osteoblastic OSA and 2 (6%) were giant cell OSA. Ten OSAs (29%) were classified as grade I, 14 (41%) as grade II and 10 (29%) as grade III. Five cases had histological evidence of intravascular neoplastic invasion. Immunohistochemistry In 24 cases (71%) a cytoplasmic expression for IGF-1R in the neoplastic cells was observed. The number of positive cells in all cases was >76% of the neoplastic cells in the section. Eight (24%) cases showed mild cytoplasmic immunolabelling (low-IGF-1R expression group; Figs. 1a and b), while 16 (47%) cases displayed strong cytoplasmic immunolabelling (high IGF-1R-expression group; Figs. 1c and d). In 10 (29%) cases immunolabelling was not observed.

Statistical analyses comparing the immunohistochemical IGF1R expression with all clinicopathological data listed in Table 1 did not show any statistical association. Univariate analysis indicated that high IGF-1R expression was a significant prognostic indicator for survival (log-rank test P < 0.05; odds ratio [OR], 2.2; 95% confidence interval [CI], 1.0–4.9), but was not an indicator for DFI (logrank test P < 0.05; Fig. 2). The other histopathological criteria examined demonstrated that poorly differentiated OSA (III grade) was a good indicator of both survival (log-rank test P < 0.00; OR, 4.2; 95% CI, 1.1–15.7) and DFI (log-rank test P < 0.001; OR, 3.5; 95% CI, 1.0–12.7). On multivariate analysis, both histological grade III (P < 0.05; OR, 4.1; 95% CI, 1.3–12.6) and high IGF-1R expression (P < 0.05; OR 2.9; 95% CI, 1.1–8.0) had predictive power with regard to OS. Multivariate analysis indicated that only histological grade III (P < 0.05; OR, 4.3; 95% CI, 1.3–14.0) had predictive power for DFI, while serum alkaline phosphatase (ALP) values did not have a prognostic significance (P = 0.054).

Molecular investigations Quantitative data obtained from q-PCR, showed a significant increase (P < 0.05) of IGF-1R transcript in Wall, Penny, Sky, Dark, Pedro and Lord cell lines compared to OSB1, while IGF-1 transcript levels were significantly decreased (P < 0.05) in Wall, Penny, Dark, Desmon, Pedro and Lord cell lines (Fig. 3). IGF-1 and IGF-1R proteins were evaluated by western blot analysis on the seven primary OSA cell lines and on OSB1. IGF-1 protein was not detectable (data not shown), while a specific band corresponding to canine IGF-1R (130 kDa) was found in all OSA cell lines except in Sky, with a higher expression in the Wall, Penny and Desmond and OSB cell lines (Fig. 4). In order to investigate which molecular pathways were correlated to IGF-1R expression, total AKT, p-AKT and p-MAPK were also investigated by western blot. Specific bands corresponding to the total form of AKT (60 kDa) were present in all OSA cell lines with similar intensity, while the activated form of AKT (p-AKT) was expressed at higher levels in the Penny and Dark cell lines and at lower levels in the OSB1 cell line. The double band corresponding to the activated form of MAPK (p-MAPK – 44 kDa) was visible in all cell lines.

Statistical analysis

Biological effects of pycropodophillin on canine osteosarcoma cell lines

Animals with tumours with high IGF-1R expression had a lower survival (mean, 279.5 days) compared to those with OSAs with low IGF-1R (423.8 days, log-rank test P < 0.05; Fig. 2) in both multivariate and univariate analyses, while DFI was reduced but not statistically different (246.8 vs. 347.7 days, respectively; Fig. 2). Moreover, dogs with low IGF-1R expression survived more than 200 days longer (Fisher’s exact test: P < 0.05) than those with high IGF-1R expression.

In vitro treatment of OSA cell lines with increasing concentration of PPP induced cell death in a dose-dependent manner. IC50s were 0.25 μM at 48 h and 0.5 μM for both 24 and 48 h (data not shown). A statistically significant increase of caspase 3/7 activity with 0.5 μM PPP after both 24 and 48 h was observed (Fig. 5) (P < 0.05), while after 48 h of incubation with 0.25 μM PPP (value corresponding to IC50), caspase activity did not significantly increase.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

5

Fig. 1. (a) Grade III osteoblastic productive osteosarcomas. Neoplastic cells are characterised by absent cytoplasmic immunolabelling for IGF-1R (negative); (b) grade III osteoblastic productive osteosarcomas. Neoplastic cells are characterised by diffuse and weak cytoplasmic immunolabelling for IGF-1R (1+; low IGF-1R expression). Streptavidinbiotin–peroxidase method. Mayer’s haematoxylin counterstaining. Scale bar, 100 μM; (c) grade II osteoblastic productive osteosarcomas. Neoplastic cells are characterised by diffuse and moderate cytoplasmic immunolabelling for IGF-1R (2+; high IGF-1R expression). Streptavidin-biotin–peroxidase method. Mayer’s haematoxylin counterstaining. Scale bar, 100 μM; (d) grade II osteoblastic productive osteosarcomas. Neoplastic cells are characterised by diffuse and strong cytoplasmic immunolabelling for IGF-1R (3+; high IGF-1R expression). Streptavidin-biotin–peroxidase method. Mayer’s haematoxylin counterstaining. Scale bar, 100 μM.

In order to investigate the biochemical effects of PPP on the Penny cell line, the expressions of IGF-1R, p-AKT and p-MAPK were measured by western blot after 24 and 48 h of treatment. In these conditions no variations in IGF-1R and p-AKT expression were detected, while a dose- and time-dependent increased expression of p-MAPK, more evident at 0.5 μM after 24 h and 0.5 μM after 48 h, was observed (Fig. 6). The effects of PPP and IGF-1 on p-IGF-1R in Penny and Sky cell lines were also evaluated: in the Penny cell line, IGF-1 alone does not change the phosphorylation status of IGF1R, while PPP dephosphorylated IGF-1R. This biological effect was not detectable in the Sky cell line that does not express IGF-1R (Fig. 7). In addition, down-regulation of p-AKT and p-MAP was observed in the Penny cell line treated with PPP (Fig. 8). Discussion OSA is the most frequently diagnosed canine bone tumour and, despite the use of aggressive treatments such as wide tumour resection and adjuvant chemotherapy, it remains a major cause of death

in dogs as in humans (Morello et al., 2011). Therefore, new therapeutic strategies based on targeted therapy could improve prognosis and therapeutic success. Among the potential targeted therapies, TKIs represent an attractive approach to treat cancer in humans and animals. Specific inhibitors against PDGFRs, C-KIT, and VEGFRs have shown (London, 2009; Rankin et al., 2012; Casteleyn et al., 2013) that TKRs, when expressed, can be prognostic factors in tumours and important molecular therapeutic targets (Yang and Zhang, 2013). This is the first study investigating IGF-1R expression in canine OSA tissues, although a previous study demonstrated IGF-1R expression in OSA cell lines (MacEwen et al., 2004). In the current study, IGF-1R was expressed in 24/34 OSAs (71%) similar to what has been described in human OSAs (86%) (Wang et al., 2012). To better understand the prognostic value of IGF1-R expression, canine OSAs with high and low IGF-1R expressions were compared in terms of OS and DFI. Dogs with high IGF-1R expression were characterised by a shorter survival when compared to those with low IGF-1R expression in both univariate and multivariate analyses, but no difference were observed for DFI.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS 6

L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

Fig. 2. (a) Kaplan–Meier curve of overall survival (OS) in dogs with low-IGF-1R expression (median OS: 423.8 days) and high-IGF-1R expression (median OS: 279.5 days) osteosarcomas (log-rank test: P < 0.05); (b) Kaplan–Meier curve of disease free interval (DFI) in dogs with low-IGF-1R expression (347.7 days) and high-IGF-IR expression osteosarcomas (246.8 days – log-rank test: P > 0.05).

The multivariate analysis also demonstrated that a poorly differentiated histological grade was the most valuable prognostic variable in this study, while ALP serum values, a well-known predictor of outcome (Garzotto et al., 2000), did not reach statistical difference, probably due to the small number of samples. Dogs with low IGF-1R expression survived more than 200 days longer than dogs with high IGF-R expression. This prognostic value is comparable to the findings observed by Wang et al. (2012), in which humans with OSA characterised by high IGF-1R expression had a lower 5-year OS rate than those with low IGF-1R.

In the current study, IGF-1 mRNA and protein expression were not observed in cell lines and tumour samples, except for the Sky cells. This may suggest that in canine OSA, IGF-1R is activated by IGF-1 mostly in a paracrine or endocrine, rather than autocrine manner (Werner et al., 2008; Werner and Bruchim, 2009). IGF-1R signalling plays a key role in bone growth and development, but aberrant signalling of this pathway has been implicated in various human cancer types, including breast (Yang and Yee, 2012), prostate (Price et al., 2012), lung (Tsuta et al., 2013) and sarcomas (Asmane et al., 2012). Moreover, its expression induces

Fig. 3. Expression by q-PCR of IGF-I and IGF-1R transcripts in primary osteosarcoma cell lines. The fold increase of each specific mRNA was normalised with normal OSB1 cell line and the error bars indicate the standard deviation of experimental triplicates. * P < 0.05 vs. OSB1.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

Fig. 4. Expression by western blot analysis of IGF-1R α. AKT, p-AKT and p-MAPK in primary OSA cell lines (Wall, Penny, Dark, Sky, Pedro, Lord, and Desmon) and in normal canine osteoblasts (OSB1). α-Tubulin served as loading control.

chemoresistance and anti-apoptotic processes (Grothey et al., 1999). In veterinary oncology, IGF receptors have been studied only in canine mammary tumours (Dolka et al., 2011; Yang and Yee, 2012), insulinoma (Buishand et al., 2012) and canine melanomas (Thamm et al., 2010). In canine mammary tumours, the high level of IGF1R expression was correlated to high oestrogen receptor (ER) alpha and progesterone receptor (PR) expression and to the expression of anti-apoptotic molecules such as p53 and Bax (Dolka et al., 2011). The Wall, Penny and Desmon cell lines overexpressed IGF-1R when compared to the OSB1 line and IGF1-R expression correlated with activation of both AKT (two cell lines) and MAPK (all cell lines). IGF-1R was not detected in Sky cells and was only marginally expressed in Dark cells. The high level of p-MAPK expression in Dark cells suggests that MAPK can be phosphorylated through IGF-1R-independent signalling. When comparing protein and mRNA data, lower mRNA expression of IGF-1R was found in Desmon cell line compared to the OSB1 line. This disagreement can be explained by the fact that protein levels do not always mirror mRNA levels: low protein degradation and turnover can result in high protein levels even in the presence of low mRNA levels. The molecular and immunohistochemical findings in the current study suggest that IGF-1R signalling could play an important role in the pathogenesis of canine OSA. Recently, PPP (Girnita et al., 2004) and a monoclonal antibody against IGF-1R (Asmane et al., 2012) have

7

Fig. 6. Evaluation of IGF-1Rα, p-AKT and p-MAPK expression in the Penny cell line after treatment with PPP used at different time periods (24 h and 48 h) and concentrations (ranging from 0.01 to 0.5 μM). α-Tubulin served as loading control.

been used to specifically inhibit IGF-1R. Several studies in humans have demonstrated the ability of PPP to block proliferation and induce apoptosis in multidrug resistant OSA cell lines (Duan et al., 2009). PPP has several advantages in that it does not react with the insulin receptor and induces very limited tumour resistance. In our study, PPP at 0.5 μM for 24 and 48 h induced caspase 3/7 activation, but not at 0.25 μM, the concentration associated with the IC50 for cytotoxicity. Likewise, PPP treatment led to a significant increase in MAPK activation, which has been shown to be involved in cell death (Martin and Pognonec, 2010; Subramaniam and Unsicker, 2010; Teixeiro and Daniels, 2010) and to represent a proapoptotic factor through ERK and caspase 3 activation (Cagnol and Chambard, 2010). This is supported by recent results, showing that neoplastic cell lines treated in vitro with TKIs show specific activation of MAPK (Schnetzke et al., 2014). Finally, PPP treatment also caused p-IGF-1R dephosphorylation with a partial blocking of p-MAPK and p-AKT, as it has been shown in humans with specific inhibition of IGF-1R. Conclusions This study demonstrated that IGF-1R most likely plays a role in the pathogenesis of canine OSA and that its expression is

Fig. 5. Caspase 3/7 activity measured by fluorescence in the Penny cell line treated with pycropodophillin (PPP) used for different times (24 and 48 h) and concentrations (ranging from 0.01 to 0.5 μM). Level of caspase 3/7 activity was reported after subtraction of fluorescence levels of wells with media only. Error bars represent standard error of two different experiments with triplicate samples. * P < 0.05 vs. t0.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

8

of Comparative Pathology Bruno Maria Zaini of the Faculty of Veterinary Medicine of Turin (Italy). Appendix: Supplementary material Supplementary data to this article can be found online at doi:10.1016/j.tvjl.2014.09.005. References

Fig. 7. Evaluation of IGF-1Rα and its phosphorylated form (p-IGF-1R) in the Penny and Sky cell lines after treatment with and without IGF-1 (1 μM) and PPP (0.5 μM). α-Tubulin served as loading control.

Fig. 8. Evaluation of IGF-1Rα, p-IGFR1, p-AKT and p-MAPK expression in the Penny cell line after treatment with and without IGF-1 (1 μM) and PPP (0.5 μM). α-Tubulin served as loading control.

correlated with a poor prognosis. Moreover, IGF1-R inhibition by PPP suggests that IGF-1R may represent a suitable therapeutic target. These data further show the high level of similarity between canine and human OSA.

Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the paper.

Acknowledgements This work was funded by the grant IDORTO1179JL from the University of Turin (Compagnia di San Paolo). The authors wish to thank Alessandra Sereno for technical support and the Reference Centre

Asmane, I., Watkin, E., Alberti, L., Duc, A., Marec-Berard, P., Ray-Coquard, I., Cassier, P., Decouvelaere, A.V., Ranchere, D., Kurtz, J.E., et al., 2012. Insulin-like growth factor type 1 receptor (IGF-1R) exclusive nuclear staining: A predictive biomarker for IGF-1R monoclonal antibody (Ab) therapy in sarcomas. European Journal of Cancer 48, 3027–3035. Avnet, S., Sciacca, L., Salerno, M., Gancitano, G., Cassarino, M.F., Longhi, A., Zakikhani, M., Carboni, J.M., Gottardis, M., Giuti, A., et al., 2009. Insulin receptor isoform A and insulin-like growth factor II as additional treatment targets in human osteosarcoma. Cancer Research 69, 2443–2452. Baserga, R., 1995. The insulin-like growth-factor-I receptor – A key to tumor-growth. Cancer Research 55, 249–252. Bohula, E.A., Playford, M.P., Macaulay, V.M., 2003. Targeting the type 1 insulin-like growth factor receptor as anti-cancer treatment. Anti-Cancer Drugs 14, 669–682. Buishand, F.O., van Erp, M.G.M., Groenveld, H.A., Mol, J.A., Kik, M., Robben, J.H., Kooistra, H.S., Kirpensteijn, J., 2012. Expression of insulin-like growth factor-1 by canine insulinomas and their metastases. The Veterinary Journal 191, 334–340. Cagnol, S., Chambard, J.C., 2010. ERK and cell death: Mechanisms of ERK-induced cell death-apoptosis, autophagy and senescence. The FEBS Journal 277, 2–21. Casteleyn, C., Sleeckx, N., De Spiegelaere, W., Heindryckx, F., Coulon, S., Van Steenkiste, C., 2013. New therapeutic targets in veterinary oncology: Man and dog definitely are best friends. The Veterinary Journal 195, 6–7. De Maria, R., Miretti, S., Iussich, S., Olivero, M., Morello, E., Bertotti, A., Christensen, J.G., Biolatti, B., Levine, R.A., Buracco, P., et al., 2009. Met oncogene activation qualifies spontaneous canine osteosarcoma as a suitable pre-clinical model of human osteosarcoma. Journal of Pathology 218, 399–408. Dolka, I., Motyl, T., Malicka, E., Sapierzynski, R., Fabisiak, M., 2011. Relationship between receptors for insulin-like growth factor-I, steroid hormones and apoptosis-associated proteins in canine mammary tumors. Polish Journal of Veterinary Sciences 14, 245–251. Duan, Z., Choy, E., Harmon, D., Yang, C., Ryu, K., Schwab, J., Mankin, H., Hornicek, F.J., 2009. Insulin-like growth factor-I receptor tyrosine kinase inhibitor cyclolignan picropodophyllin inhibits proliferation and induces apoptosis in multidrug resistant osteosarcoma cell lines. Molecular Cancer Therapeutics 8, 2122–2130. Farese, J.P., Kirpensteijn, J., Kik, M., Bacon, N.J., Waltman, S.S., Seguin, B., Kent, M., Liptak, J., Straw, R., Chang, M.N., et al., 2009. Biologic behavior and clinical outcome of 25 dogs with canine appendicular chondrosarcoma treated by amputation: A Veterinary Society of Surgical Oncology retrospective study. Veterinary Surgery 38, 914–919. Flint, A.E., U’Ren, L., Legare, M.E., Withrow, S.J., Dernell, W., Hanneman, W.H., 2004. Overexpression of the erbB-2 proto-oncogene in canine osteosarcoma cell lines and tumors. Veterinary Pathology 41, 291–296. Fossey, S.L., Liao, A.T., McCleese, J.K., Bear, M.D., Lin, J., Li, P.-K., Kisseberth, W.C., London, C.A., 2009. Characterization of STAT3 activation and expression in canine and human osteosarcoma. BMC Cancer 9, 81. Garzotto, C.K., Berg, J., Hoffmann, W.E., Rand, W.M., 2000. Prognostic significance of serum alkaline phosphatase activity in canine appendicular osteosarcoma. Journal of Veterinary Internal Medicine 14, 587–592. Girnita, A., Girnita, L., del Prete, F., Bartolazzi, A., Larsson, O., Axelson, M., 2004. Cyclolignans as inhibitors of the insulin-like growth factor-1 receptor and malignant cell growth. Cancer Research 64, 236–242. Giustina, A., Mazziotti, G., Canalis, E., 2008. Growth hormone, insulin-like growth factors, and the skeleton. Endocrine Reviews 29, 535–559. Grothey, A., Voigt, W., Schober, C., Muller, T., Dempke, W., Schmoll, H.J., 1999. The role of insulin-like growth factor I and its receptor in cell growth, transformation, apoptosis, and chemoresistance in solid tumors. Journal of Cancer Research and Clinical Oncology 125, 166–173. Johnson, A.S., Couto, C.G., Weghorst, C.M., 1998. Mutation of the p53 tumor suppressor gene in spontaneously occurring osteosarcomas of the dog. Carcinogenesis 19, 213–217. Klopfleisch, R., Hvid, H., Klose, P., da Costa, A., Gruber, A.D., 2010. Insulin receptor is expressed in normal canine mammary gland and benign adenomas but decreased in metastatic canine mammary carcinomas similar to human breast cancer. Veterinary and Comparative Oncology 8, 293–301. Kubota, T., Elalieh, H.Z., Saless, N., Fong, C., Wang, Y., Babey, M., Cheng, Z., Bikle, D.D., 2013. Insulin-like growth factor-1 receptor in mature osteoblasts is required for periosteal bone formation induced by reloading. Acta Astronautica 92, 73–78. Lana, S.E., Ogilvie, G.K., Hansen, R.A., Powers, B.E., Dernell, W.S., Withrow, S.J., 2000. Identification In of matrix metalloproteinases in canine neoplastic tissue. American Journal of Veterinary Research 61, 111–114. Levine, R.A., Forest, T., Smith, C., 2002. Tumor suppressor PTEN is mutated in canine osteosarcoma cell lines and tumors. Veterinary Pathology 39, 372–378. London, C.A., 2009. Tyrosine kinase inhibitors in veterinary medicine. Topics in Companion Animal Medicine 24, 106–112.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005

ARTICLE IN PRESS L. Maniscalco et al./The Veterinary Journal ■■ (2014) ■■–■■

Loukopoulos, P., Robinson, W.E., 2007. Clinicopathological relevance of tumour grading in canine osteosarcoma. Journal of Comparative Pathology 136, 65–73. Loukopoulos, P., Mungall, B.A., Straw, R.C., Thornton, J.R., Robinson, W.F., 2003. Matrix metalloproteinase-2 and-9 involvement in canine tumors. Veterinary Pathology 40, 382–394. MacEwen, E.G., Kutzke, J., Carew, J., Pastor, J., Schmidt, J.A., Tsan, R., Thamm, D.H., Radinsky, R., 2003. C-Met tyrosine kinase receptor expression and function in human and canine osteosarcoma cells. Clinical and Experimental Metastasis 20, 421–430. MacEwen, E.G., Pastor, J., Kutzke, J., Tsan, R., Kurzman, I.D., Thamm, D.H., Wilson, M., Radinsky, R., 2004. IGF-1 receptor contributes to the malignant phenotype in human and canine osteosarcoma. Journal of Cellular Biochemistry 92, 77–91. Maniscalco, L., Iussich, S., Morello, E., Martano, M., Biolatti, B., Riondato, F., Della Salda, L., Romanucci, M., Malatesta, D., Bongiovanni, L., et al., 2013. PDGFs and PDGFRs in canine osteosarcoma: New targets for innovative therapeutic strategies in comparative oncology. The Veterinary Journal 195, 41–47. Martin, P., Pognonec, P., 2010. ERK and cell death: Cadmium toxicity, sustained ERK activation and cell death. The FEBS Journal 277, 39–46. Morello, E., Martano, M., Buracco, P., 2011. Biology, diagnosis and treatment of canine appendicular osteosarcoma: Similarities and differences with human osteosarcoma. The Veterinary Journal 189, 268–277. Price, A.J., Allen, N.E., Appleby, P.N., Crowe, F.L., Travis, R.C., Tipper, S.J., Overvad, K., Gronbaek, H., Tjonneland, A., Johnsen, N.F., et al., 2012. Insulin-like growth factor-I concentration and risk of prostate cancer: Results from the European prospective investigation into cancer and nutrition. Cancer Epidemiology Biomarkers and Prevention 21, 1531–1541. Rainusso, N., Brawley, V.S., Ghazi, A., Hicks, M.J., Gottschalk, S., Rosen, J.M., Ahmed, N., 2012. Immunotherapy targeting HER2 with genetically modified T cells eliminates tumor-initiating cells in osteosarcoma. Cancer Gene Therapy 19, 212–217. Rankin, K.S., Starkey, M., Lunec, J., Gerrand, C.H., Murphy, S., Biswas, S., 2012. Of dogs and men: Comparative biology as a tool for the discovery of novel biomarkers and drug development targets in osteosarcoma. Pediatric Blood and Cancer 58, 327–333. Schnetzke, U., Fischer, M., Frietsch, J.J., Finkensieper, A., Clement, J.H., Hochhaus, A., La Rosée, P., 2014. Paradoxical MAPK-activation in response to treatment with tyrosine kinase inhibitors in CML: Flow cytometry loses track. Cytometry. Part B, Clinical Cytometry 86, 229–235. Slayter, M.V., Boosinger, T.R., Pool, R.R., Dammrich, K., Misdorp, W., Larsen, S., 1994. WHO Histological Classification of Bone and Joint Tumors of Domestic Animals, Second Ed. Armed Forces Institute of Pathology, Washington, DC.

9

Subramaniam, S., Unsicker, K., 2010. ERK and cell death: ERK1/2 in neuronal death. The FEBS Journal 277, 22–29. Tada, H., Shiho, O., Kuroshima, K., Koyama, M., Tsukamoto, K., 1986. An improved colorimetric assay for interleukin 2. Journal of Immunology Methods 93, 157–165. Teixeiro, E., Daniels, M.A., 2010. ERK and cell death: ERK location and T cell selection. The FEBS Journal 277, 30–38. Thamm, D.H., Huelsmeyer, M.K., Mitzey, A.M., Ourollo, B., Rose, B.J., Kurzman, I.D., 2010. RT-PCR-based tyrosine kinase display profiling of canine melanoma: IGF-1 receptor as a potential therapeutic target. Melanoma Research 20, 35– 42. Tsuta, K., Mimae, T., Nitta, H., Yoshida, A., Maeshima, A.M., Asamura, H., Grogan, T.M., Furuta, K., Tsuda, H., 2013. Insulin-like growth factor-1 receptor protein expression and gene copy number alterations in non small cell lung carcinomas. Human Pathology 44, 975–982. Vasilcanu, D., Weng, W.H., Girnita, A., Lui, W.O., Vasilcanu, R., Axelson, M., Larsson, O., Larsson, C., Girnita, L., 2006. The insulin-like growth factor-1 receptor inhibitor PPP produces only very limited resistance in tumor cells exposed to long-term selection. Oncogene 25, 3186–3195. Wang, Y.-H., Han, X.-D., Qiu, Y., Xiong, J., Yu, Y., Wang, B., Zhu, Z.-Z., Qian, B.-P., Chen, Y.-X., Wang, S.-F., et al., 2012. Increased expression of insulinlike growth factor-1 receptor is correlated with tumor metastasis and prognosis in patients with osteosarcoma. Journal of Surgical Oncology 105, 235–243. Werner, H., Bruchim, I., 2009. The insulin-like growth factor-I receptor as an oncogene. Archives of Physiology and Biochemistry 115, 58–71. Werner, H., Weinstein, D., Bentov, I., 2008. Similarities and differences between insulin and IGF-I: Structures, receptors, and signalling pathways. Archives of Physiology and Biochemistry 114, 17–22. Withrow, S.J., Powers, B.E., Straw, R.C., Wilkins, R.M., 1991. Comparative aspects of osteosarcoma: Dogs versus man. Clinical Orthopaedics and Related Research 270, 159–168. Wu, X., Sooman, L., Wickstrom, M., Fryknas, M., Dyrager, C., Lennartsson, J., Gullbo, J., 2013. Alternative cytotoxic effects of the postulated IGF-1R inhibitor picropodophyllin (PPP) in vitro. Molecular Cancer Therapeutics 12, 1526–1536. Yang, J.L., Zhang, W., 2013. New molecular insights into osteosarcoma targeted therapy. Current Opinion in Oncology 25, 398–406. Yang, Y.Z., Yee, D., 2012. Targeting insulin and insulin-like growth factor signaling in breast cancer. Journal of Mammary Gland Biology and Neoplasia 17, 251–261. Zhang, W., Shen, X., Wan, C., Zhao, Q., Zhang, L., Zhou, Q., Deng, L., 2012. Effects of insulin and insulin-like growth factor 1 on osteoblast proliferation and differentiation: Differential signalling via Akt and ERK. Cell Biochemistry and Function 30, 297–302.

Please cite this article in press as: Lorella Maniscalco, Selina Iussich, Emanuela Morello, Marina Martano, Francesca Gattino, Silvia Miretti, Bartolomeo Biolatti, Paolo Accornero, Eugenio Martignani, Raquel Sánchez-Céspedes, Paolo Buracco, Raffaella De Maria, Increased expression of insulin-like growth factor-1 receptor is correlated with worse survival in canine appendicular osteosarcoma, The Veterinary Journal (2014), doi: 10.1016/j.tvjl.2014.09.005