TYPE Original Research PUBLISHED 30 November 2023 DOI 10.3389/fvets.2023.1253074 Evaluation of in vitro intrinsic OPEN ACCESS radiosensitivity and characterization of five canine EDITED BY Isaac Karimi, Razi University, Iran REVIEWED BY Sam Long, high-grade glioma cell lines Veterinary Referral Hospital, Australia Hidetaka Nishida, Azabu University, Japan Benjamin Cartiaux 1,2*, Alexandra Deviers 2, Caroline Delmas 1,3, *CORRESPONDENCE Jérôme Abadie 4, Martí Pumarola Battle 5, Benjamin Cartiaux Elizabeth Cohen-Jonathan Moyal 1,3 and Giovanni Mogicato 2 benjamin.cartiaux@envt.fr 1 INSERM UMR.1037-Cancer Research Center of Toulouse (CRCT), University Paul Sabatier Toulouse III, RECEIVED 04 July 2023 Toulouse, France, 2 ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, ENVT, ACCEPTED 17 November 2023 Toulouse, France, 3 IUCT-oncopole, Toulouse, France, 4 Department of Biology, Pathology and Food PUBLISHED 30 November 2023 Sciences, Laboniris, Nantes, France, 5 Unit of Murine and Comparative Pathology, Department of Animal CITATION Medicine and Surgery, Veterinary Faculty, Autonomous University of Barcelona, Barcelona, Spain Cartiaux B, Deviers A, Delmas C, Abadie J, Pumarola Battle M, Cohen-Jonathan Moyal E and Mogicato G (2023) Evaluation of in Glioma is the most common primary brain tumor in dogs and predominantly vitro intrinsic radiosensitivity and characterization of five canine high-grade affects brachycephalic breeds. Diagnosis relies on CT or MRI imaging, and the glioma cell lines. proposed treatments include surgical resection, chemotherapy, and radiotherapy Front. Vet. Sci. 10:1253074. depending on the tumor’s location. Canine glioma from domestic dogs could doi: 10.3389/fvets.2023.1253074 be used as a more powerful model to study radiotherapy for human glioma than COPYRIGHT the murine model. Indeed, (i) contrary to mice, immunocompetent dogs develop © 2023 Cartiaux, Deviers, Delmas, Abadie, Pumarola Battle, Cohen-Jonathan Moyal and spontaneous glioma, (ii) the canine brain structure is closer to human than mice, Mogicato. This is an open-access article and (iii) domestic dogs are exposed to the same environmental factors than distributed under the terms of the Creative humans. Moreover, imaging techniques and radiation therapy used in human Commons Attribution License (CC BY). The use, distribution or reproduction in other medicine can be applied to dogs, facilitating the direct transposition of results. forums is permitted, provided the original The objective of this study is to fully characterize 5 canine glioma cell lines and author(s) and the copyright owner(s) are to evaluate their intrinsic radiosensitivity. Canine cell lines present numerous credited and that the original publication in this journal is cited, in accordance with accepted analogies between the data obtained during this study on different glioma cell academic practice. No use, distribution or lines in dogs. Cell morphology is identical, such as doubling time, clonality test and reproduction is permitted which does not karyotype. Immunohistochemical study of surface proteins, directly on cell lines comply with these terms. and after stereotaxic injection in mice also reveals close similarity. Radiosensitivity profile of canine glial cells present high profile of radioresistance. KEYWORDS glioma, canine, characterization, translational, radiation 1 Introduction The increase of canine life expectancy, linked to improved nutrition, better vaccination, and enhanced veterinary care, is associated with a rise in age-related health conditions such as cancer. In fact, in the United States, more than 1 million cases of cancer are diagnosed in dogs each year (1) Among brain tumors in dogs, gliomas are the most common primary intracranial tumors, accounting for 36 to 70% of cases according to the literature (2). Brachycephalic dogs are the most commonly affected; among them, Boston Terriers, Boxers, and English and French Bulldogs are the most susceptible (3). It seems that genes predisposing to the development of gliomas exist and have been preserved through selective breeding of certain brachycephalic breeds. The majority of gliomas in dogs are localized in the cerebral hemispheres, preferably in Frontiers in Veterinary Science 01 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 the frontal lobe, the temporal lobe, and the parietal lobe (4). Clinical the growth properties of cells and the proportion of tumorigenic cells, signs associated with gliomas in dogs are similar to those identified in we have also evaluated the general and phenotypic characteristics of humans. The most frequently encountered sign is the onset of seizures these cell lines and assessed their tumorigenicity. in nearly 50% of dogs. These seizures are a direct result of the mass effect exerted by the tumor on brain tissue, or they are secondary to peritumoral edema or neuroinflammation induced by the glioma. 2 Materials and methods Dogs may also exhibit other nonspecific clinical signs such as lethargy, inappetence, or weight loss (5). The MRI appearance of canine gliomas 2.1 Cell culture tends to exhibit a similar pattern to what is observed in humans: they appear as hyperintense masses on T2 and iso- to hypo-intense on T1, Canine cell lines were generously provided by Dr. Michael Berens with no contrast enhancement for low-grade gliomas and variable (J3T), Dr. Peter Dickinson (J3T-Bg, SDT-3G, G06A) and Dr. Jérôme contrast enhancement for high-grade gliomas (6). The treatment of Abadie (Raffray). The J3T cell line was derived from an anaplastic gliomas in dogs is diverse and relies on various methods such as astrocytoma in a 10 year-old male Boston Terrier (37). The J3T-Bg surgery, chemotherapy, and radiotherapy. The latter approach has glioma cell line was derived from the canine J3T cell line following shown promising results in the treatment of gliomas in dogs, with an passage as a subcutaneous tumor through Biege-Nude-xid mice (16). increase in median survival (7). The SDT3G and G06A cell lines were derived from glioblastomas in a Pet dogs with spontaneously arising high-grade gliomas are 12 year-old male English Bulldog and a 2-year old female becoming widely acknowledged as relevant translational models that ovariohysterectomized Australian Shepherd, respectively (17). Raffray could be used in addition to murine models. Indeed, the spontaneous cell line was derived from an anaplastic oligodendroglioma in a oncogenesis of canine glioma is inherently associated with inter- 10-year-old female Boxer. All cell lines were grown in T-25 cm2 flasks tumor heterogeneity, genetic variability and anti-tumoral immune with Dulbecco’s Modified Eagle Medium (Gibco™), supplemented response that are lacking in immunocompromised xenograft mouse with 10% fetal bovine serum, 1% penicillin/streptomycin models. Canine high-grade gliomas closely resemble their human (Gibco-BRL). Cells were maintained in an incubator at 37°C counterparts regarding oncogenic mechanisms (8–10), cancer stem with 5% CO2. cells (11–13), histopathological features (14), molecular alterations (15–19) and immune microenvironment (9, 20). Furthermore, the large size of their brain enables the use of the same RT devices and 2.2 Cell proliferation MRI scanners that those used in human medicine (21, 22). Taken all together, these data suggest that glioma-bearing dogs could be a well- Cells of each cell line were seeded at a concentration of 1×105 cells suited pre-clinical model to predict the efficacy of radiosensitizers in per well in 6-well plates and maintained in supplemented DMEM human patients. (10% fetal bovine serum, 1% penicillin/streptomycin) at 37°C for Even if the clinical response to RT is described in literature (21), 5 days. Every 24 h, two wells were trypsinized for 5 min at 37°C and to date there has been no published work evaluating the radiobiologic the cells were counted using a Thomas cell counting chamber after a parameters of canine high-grade gliomas. These parameters are dilution 1: 1 with Trypan Blue. The average was carried out between obtained in vitro by performing clonogenic assays which are the two wells. Replicates from three 6-well plates were scored each day. considered as the gold standard for determining reproductive tumor The doubling times of each cell line were computed using the “cell cell death induced by ionizing radiation (IR). The surviving fraction calculator++” tool (doubling-time. Com; Roth V., 2006). at 2 Gy (SF2), the fraction of cells surviving a single 2 Gy dose of IR, and the mean inactivation dose (MID) are parameters that can be derived from clonogenic assays. SF2 and MID are useful tools to 2.3 Chromosome number study the intrinsic radiosensitivity of cell lines and to predict in vivo tumor sensitivity to radiation (23–25). So far, four canine high-grade Metaphase cells were harvested by physical shaking and glioma cell lines have been put forward in the literature: J3T, J3T-Bg, centrifuged at 500 g for 10 min, then they were resuspended in SDT-3G, and G06A. The J3T cells were used to develop animal hypotonic solution of fetal calf serum (1:6) for 20 min, at 37°C. Cells models of canine gliomas (26–28), and all of these cell lines have were then centrifuged at 500 g for 10 min and fixed in ethanol: acetic proven useful in understanding the mechanisms of oncogenesis in acid (3,1) for at least an hour, at 4°C. Cells were spread on a slide and canine high-grade gliomas (8, 16, 17, 29), evaluating their sensitivity stained with 1% Giemsa solution to generate GTG banding. At least to chemotherapeutics (17, 30), targeted therapies (31–34) and viral 20 metaphases were evaluated for each cell line to count the total vectors (27), as well as examining their miRNA profile in response number of chromosomes and the number of metacentric to hypoxia (35). Recently, a new glioma cell line called Raffray, chromosomes per cell. derived from a tumor originally classified as an anaplastic oligodendroglioma in a 7-year-old Boxer, was developed in a French laboratory (36). 2.4 Orthotopic xenograft generation In this context, the aim of the present study was to evaluate the intrinsic radiation sensitivity of these five canine high-grade glioma Orthotopic canine high-grade xenografts were established in cell lines by performing clonogenic assays and examining the SF2 8 week-old female nude mice (Janvier Labs) as previously described value and MID. As the tumor response to irradiation is dependent on (38). Briefly, mice were anesthetized with intraperitoneal ketamine Frontiers in Veterinary Science 02 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 (100 mg/kg) and xylazine (10 mg/kg) and received a stereotaxically antibody labeling and detection were performed as previously guided injection of 2.5 × 105 cells suspended in 5 μL DMEM, using a described (12). Primary antibodies, dilution and pre-treatment are Hamilton syringe. The injection was precisely located into the right summarized in Supplementary Table S1. The positive controls used forebrain (2 mm lateral and 1 mm anterior to the bregma at a 5 mm for GFAP, Olig2, S-100, Vim, NeuN, and DCx immunostaining of depth from the skull surface). The mice were observed for behavior xenogafts were samples of normal brain including grey and white changes suggestive of brain tumor (head tilting, circling, tremors) or matter, and skin for Vim immunostaining. Positive control for general impairment of health (hypoactivity, decreased body weight of CD133 was adult canine healthy kidney tissue, because it is expressed more than 20%, tachypnea, signs of pain such as aggressiveness, by tubular epithelial cells. In all experiments, negative controls were vocalization). When such changes were detected, they were humanely obtained by omitting the primary antibody. A proportion score euthanized by anesthetic overdose by intraperitoneal injection of (corresponding to the percentage of positive neoplastic cells in each pentobarbital confirmed by cervical dislocation. Otherwise, mice were sample) and an intensity score (corresponding to the labeling followed for behavior changes for a year. After euthanasia, brains were intensity of positive neoplastic cells in each sample) were evaluated collected for subsequent histological and immunohistochemical (12, 40). The percentage of positive neoplastic cells is obtained by analysis. All animals were treated according to the European taking the ratio of the number of labeled positive tumor cells to the Communities Council directive (2010/63/EU) regarding the care and total number of cells present in the field. Ten high-power fields were use of animals, and all the experimental procedures were approved by evaluated for each marker. An IHC score was then calculated the Institution animal ethics committee with authorization (Table 1). number 2018031614514282. 2.6 Radiosensitivity 2.5 Histo-cyto-pathology and immunohistochemistry Radiosensitivity was evaluated with a clonogenic survival assay. Canine high-grade glioma cell lines were dissociated, plated in Canine high-grade glioma cells cultured in supplemented T-25 cm2 flasks (1,500 cells/well) and irradiated at escalating doses (0, DMEM were scraped off with a sterile scalpel and transferred into a 1, 2, 4, 6, 8, and 10 Gy) with the Gamma-cell Exactor 40 (Nordion, 15 mL centrifuge tube. They were centrifuged at 400 rpm for 5 min, Ottawa, ON, Canada), as previously described (41). Irradiated cells then resuspended in 10 mL phosphate-buffered saline (PBS) after were subsequently incubated at 37°C for 15 days, to allow colony cell removal of the supernatant. This operation was repeated one more formation. Then, flasks were rinsed with 0.9% NaCl, fixed with FAA time. The resulting cell suspension was centrifuged at 400 rpm for (Formalin – Acetic acid – Alcohol) fixative, and stained by 0.1% 5 min and resuspended in 1 mL of 70% ethanol after removal of the crystal violet. Each colony with more than 50 cells was considered as supernatant. The cell suspension was then transferred in a 1.5 mL survivor. To calculate the surviving fraction at a given dose, the microcentrifuge tube and centrifuged at 400 rpm for 5 min. Once the number of colonies is divided by the number of seeded cells and supernatant was discarded, the resulting pellet was resuspended in normalized to the plating efficiency of the not irradiated controls. The 50 μL of liquefied HistoGel™ (Thermo Scientific Richard-Allan plating efficiency of not irradiated controls was determined as: Scientific, Kalamazoo, MI, United States) and the cell suspension (number of colonies formed/ number of cells inoculated) × 100. At was gently mixed to distribute cells evenly within the gel matrix least three independent experiments were carried out, from which a prior to solidification (39). HistoGel-encapsulated cells were survival curve was drawn. The linear-quadratic (LQ) model is usually transferred in pathology cassettes, fixed in 10% neutral buffered used to fit survival curves and to estimate several parameters of formalin and processed as usual before embedding in paraffin. interest: (i) SF2 as the surviving fraction of tumor cells after a dose of Excised brains were also fixed in 10% neutral buffered formalin. 2Gy; (ii) MID (mean inactivation dose) as the average dose of the Paraffin-embedded excised brains and HistoGel-encapsulated cells differential survival probability distribution; (iii) alpha (the linear were sectioned (5 μm) and stained with haematoxylin and eosin component of the LQ model) that describes cell death due to lethal (HE) for microscopic evaluation. Immunohistochemistry (IHC) was damage by a single incident; (iv) beta (the quadratic component of performed in collaboration with the laboratory team of Dr. Martí the LQ model) that represents cell death due to an accumulation of Pumarola (12). Several IHC markers were used: CD133 as a stem cell damage; and (v) the value of the alpha/beta ratio is inversely marker, Olig2 protein and doublecortin (DCx) as glial and neuronal proportional to the sensitivity to fractionation of the biological effect progenitor cell markers respectively; glial fibrillary acidic protein considered (23, 42). (GFAP) and vimentin (Vim) as mature astrocyte markers; S-100 protein as a mature oligodendroglial and astrocytic marker and βIII- tubulin and NeuN protein as mature neuron markers. 2.7 Statistical analysis Immunohistochemistry was performed in both HistoGel- encapsulated cells (DCx, GFAP, Vim, βIII-tubulin, and NeuN) and All results are represented as mean +/− standard deviation (SD). excised brains of mice which have developed tumors (Olig2, GFAP, For statistical analysis, XLSTAT software (Data Analysis and Statistical Vim, S100, NeuN, and CD133). Olig2 expression was evaluated in Solution for Microsoft Excel, Addinsoft, Paris, France 2017). Kaplan– all the xenograft samples and the other markers were evaluated only Meier estimate is used to evaluate the overall survival of mice with in two xenografts per tumor group. Procedures relative to antigen orthotopic xenografts of canine glioma cell lines. Log-rank test was retrieval, endogenous peroxidase and non-specific binding blocking, assessed to analyze significant differences between the overall survival Frontiers in Veterinary Science 03 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 TABLE 1 Evaluation of the immunohistochemical score. Proportion Percentage of Intensity score Labeling intensity of Immunohistochemical PSxIS score (PS) positive neoplastic (IS) positive neoplastic cells score cells 0 <5% 0 No labeling Negative 0 1 5–30% 1 Weak + From 1 to 3 2 30–60% 2 Mild ++ From 4 to 6 3 60–90% 3 Intense +++ From 7 to 9 4 > 90% − − ++++ From 10 à 12 TABLE 2 General genotypic and phenotypic characteristics of the canine high-grade glioma cell lines. J3T J3T-Bg SDT3 G06A Raffray DT (hours) 20.6 25.8 25.6 22.4 55.9 PE (%) 21.0 19.7 21.6 23.0 16.4 Number of chromosomes 78 to 82 78 to 81 59 to 61 78 to 82 76 to 81 Number of abnormal metacentric chromosomes 2 2 NE 1 2 GFAP ++ + Neg + + Vim +++ ++++ +++ ++++ ++++ DCx Neg Neg Neg Neg Neg βIII tubulin Neg Neg Neg Neg Neg NeuN Neg Neg Neg Neg Neg DT, doubling time; PE, plating efficiency; NE, not evaluated. The number of metacentric chromosomes for the SDT3 cell line was difficult to assess due to important chromosomal rearrangements. of different groups of cell lines. Results with p value <0.05 were were negative to GFAP. Neuronal markers (DCx, NeuN, and βIII- considered statistically significant. tubulin) were negative for all cell lines. Taken all together, the immunohistochemical results were consistent with astrocytic cell lines. Immunohistochemical results obtained for the mature cell 3 Results markers GFAP, Vimentin, and NeuN are displayed in Figure 1 for the cell line J3T-Bg. 3.1 General characteristics of the canine glioma cell lines 3.2 Radiosensitivity General characteristics of each cell line were established by assessing doubling time, plating efficiency, karyotype as well as cell The values of dose response curve derived-parameters (SF2, α, β, morphology and protein expression profile (Table 2). Doubling times α/β ratio, and MID) calculated for each cell line are summarized in were in the same range (21 to 26 h) for all the cell lines except Raffray Table 3. MID and SF2 were comparable for the 5 cell lines as cells (56 h). The determination of plating efficiency, comprised exemplified by the low coefficient of variation (CV): the mean MID between 16.4% (Raffray) and 23% (G06A), demonstrated that all was 4.84 +/− 0.64 Gy (CV: 13%) and the mean SF2 was 0.84 +/− 0.08 glioma cell lines were able to give rise to colonies, which was the (CV:10%) which is higher than the mean SF2 of cell lines derived from necessary condition to evaluate their radiosensitivity by in vitro human tumors refractory to RT (43). The mean α was 0.027 Gy−1 +/− clonogenic assay. Karyotyping revealed a number of chromosomes 0.029 (CV: 107%) and the mean α / β ratio was 0.836 +/−0.872 per cell slightly higher than the normal diploid chromosomal count (CV: 104%). for canines (78) and the presence of 1 to 2 abnormal metacentric chromosomes for all the cell lines but SDT3 which exhibited a chromosome number comprised between 59 and 61 and important 3.3 Tumorigenicity chromosomal rearrangements. Cell morphology and protein expression profile were similar for All the canine cell lines proved to be tumorigenic after orthotopic J3T, J3T-Bg, G06A, and Raffray cells: they appeared mainly as spindle- implantation in immunocompromised mice, with a tumor take rate to star-shaped cells with obvious cytonuclear atypia when observed in varying from 25% for G06A cell line to 100% for J3T and J3T-Bg cell HE-stained slides and expressed the astrocytic markers lines (Table 4). The number of mice used in experiments varies GFAP (low immunohistochemical score) and Vimentin (high between 9 and 12, depending on the cell lines. To practice the surgical immunohistochemical score, Table 2). SDT-3 cells were smaller and procedure, two unused mice from another study were added to the more rounded, they showed a strong positive reaction to Vimentin but initial group of 10 mice from J3T cell lines. Unfortunately, before Frontiers in Veterinary Science 04 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 TABLE 3 In vitro cell survival curve derived parameters of the canine high-grade glioma cell lines. J3T J3T-Bg SDT3 G06A Raffray α (Gy−1) 0.072 0.029 0.001 0.031 0.002 β (Gy−2) 0.033 0.032 0.025 0.030 0.030 α / β ratio (Gy) 2.182 0.906 0.040 1.033 0.067 SF2 (%) 73 91 93 85 80 MID (Gy) 4 4.6 5.7 4.7 5.2 FIGURE 1 Immunohistochemical features of J3T-Bg cell line embedded in HistoGel™. (A) GFAP, weak cytoplasmic immunolabeling of 60 to 90% of cells. (B) Vimentin, strong cytoplasmic staining of more than 90% of cells. (C) NeuN, negative. Scale bar: 50 μm. intracerebral injections, three mice passed away, with one from each of either (i) interlacing or parallel fascicles of eosinophilic spindle cells group used for J3T-Bg, SDT3, and G06A cell lines. with elongated nucleus, (ii) sheets and small packets of large polygonal The median survival of canine glioma-derived mouse orthotopic cells with a prominent vesicular nucleus, and (iii) a combination of xenografts was significantly different depending on the cell line that both. Tumor cells were also either spindle-shaped or polygonal in was injected in the brain (Figure 2). The median overall survival was perivascular cuffs (Figures 3D–G). These xenografts were assigned as the highest for G06A (216 days, 95% CI: 212–218) and for Raffray- high-grade tumors on the basis of the mitotic rate and the presence of derived mouse orthotopic xenografts (186 days, 95% CI:162–210). In geographic necrosis with peripheric palisading in two J3T, two J3T-Bg these two groups, mice were still alive at the endpoint of experiment and one Raffray-derived tumors (Figure 3J). Numerous branching (represented by white circles in Figure 2). The lowest median survival thin-walled capillary vessels were consistently seen in the tumors but was 22 days (20–23) for J3T-Bg. microvascular proliferation could not be observed in any of the Histological examination of xenografts are displayed in Figure 3 samples (Figure 3I). and revealed densely cellular neoplasms with different growth patterns The results of immunohistochemical results obtained for the defined as: (i) focal infiltration (well-demarcated neoplasms with focal different xenografts are displayed in Supplementary Table S2 and or multifocal regions of infiltration) for 100% of SDT3- and Figure 4. Regarding J3T, SDT3, Raffray and G06-A-derived xenografts, G06-derived xenografts and 30% of J3T-derived xenografts, (ii) diffuse Olig2 immunostaining confirmed the diagnosis of diffuse gliomas as infiltration (poorly demarcated neoplasms with a significant 79% of these tumors exhibited a high IHC score for this marker. The infiltration of the surrounding parenchyma and/or leptomeninges) for absence of expression of the NeuN marker in these tumors 10% of J3T-derived xenografts and (iii) diffuse infiltration associated corroborated their glial differentiation. All the J3T, SDT3 and Raffray- with striking perivascular cuffing for 60% of J3T tumors and 100% of derived tumors were positive to GFAP with a medium IHC score, Raffray and J3T-Bg derived xenografts (Figures 3A–C). In J3T-Bg confirming their astrocytic origin. The absence of GFAP tumors, perivascular cuffs were large and coalescent leading to the immunolabeling in G06A xenografts (Figure 4D) was consistent with formation of a multinodular mass at the injection site (Figure 3C). the oligodendroglial phenotype observed in H&E-stained sections. With regard to architectural patterns, cell morphology, cellular atypia The value of the glial marker S100 in the characterization of xenografts and mitotic rate, all the xenografts were morphologically consistent appeared limited in our study as the glial tumors with a high IHC with diffuse gliomas. G06A xenografts exhibited features of high- score for Olig2 exhibited a weak or absent S100 immunostaining. grade oligodendrogliomas. They were composed of sheets of round to Neoplastic cells did not express Vimentin in any of the tumor cores polygonal cells with a central nucleus, varying from round and except that of J3T xenografts, but a weak immunopositivity to this hyperchromatic to irregular with loose chromatin, surrounded by a marker was observed in J3T, Raffray and J3T-Bg perivascular cuffs. moderately abundant pale cytoplasm (Figure 3H). The mitotic rate CD133 immunolabeling was negative in all the samples analyzed. was comprised between 1 and 3 mitoses per high-power-field. Regarding J3T-Bg-derived xenografts, the tumor core did not express Xenografts derived from J3T, J3T-Bg, SDT3 and Raffray cells were any of the markers; the only observable immunostaining was that of suggestive of an astrocytic differentiation. Their core was composed Vimentin in perivascular cuffs (Figure 4F). Despite the absence of Frontiers in Veterinary Science 05 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 TABLE 4 Tumorigenicity of the canine high-grade glioma cell lines. J3T J3T-Bg SDT3 G06A Raffray Number of mice with intracerebral injection of tumor cells 12 9 9 9 10 Number of brains submitted to microscopic examination 12 9 9 8* 8 Number of brains harboring a tumor 12 9 8 2 4 Tumor take rate 100% 100% 89% 25% 50% Median survival (in days) 129 22 115 216 186 * Due to marked autolysis, three brains could not be analyzed (1 brain from the G06A group and two brains from the Raffray group). FIGURE 2 Kaplan–Meier analysis of overall survival in canine glioma-derived mouse orthotopic xenografts. The median survival of canine glioma-derived mouse orthotopic xenografts was significantly different depending on the cell line. The lowest median overall survival of 22 days (20–23) was observed for J3T-Bg, and the highest concerned G06A and Raffray cell lines with, respectively, 216 and 186 days. White circles represent surviving mice at the end of experiment. Olig2 labeling, J3TBg tumors were assigned as diffuse gliomas with the literature relative to J3T and Raffray cells. As described in the considering their close morphology with J3T tumors (confirmed to study of Berens et al., we observed GFAP immunolabeling of more be high-grade astrocytomas) and the presence of typical palisading than 90% of J3T cells (37). In accordance with the study of Monod, necrotic areas in 2 tumors. Because of their lack of immunopositivity Raffray cells demonstrated a weak and infrequent immunopositivity to different glial markers, they were considered to be poorly to GFAP associated with a strong Vimentin immunostaining of more differentiated gliomas. than 90% of cells (36). Although the parent tumor of Raffray cells was diagnosed as a grade III oligodendroglioma, the phenotypic characteristics of the cultured cells suggest an astrocytic differentiation. 4 Discussion The doubling times of canine glioma cells were found to be similar to those described in the literature for human GBM cell lines (from 24 The aim of the present study was to evaluate the intrinsic radiation to 60 h). The doubling time of J3T was shorter in our study compared sensitivity of five canine high-grade glioma (HGG) cell lines by to that reported in the literature (20.6 versus 49 in Berens’ study and performing clonogenic assays and examining the SF2 value and 28 in Rainov’s study). MID. To our knowledge, tumorigenicity and intrinsic radiosensitivity The radiosensitivity of canine cell lines was evaluated with a of canine HGG has never been assessed. clonogenic assay. This test, which is the most used for the evaluation Before assessing their radiosensitivity, the 5 cell lines were of the response to ionizing radiation, provide experimental survival confirmed to be neoplastic (cells exhibiting chromosomal curves representing the survival fraction (SF) as a function of abnormalities and severe cytonuclear atypia), astrocytic (cells radiation dose (44). Although the ability of SF2 to predict the immunopositive to Vimentin and GFAP and immunonegative to clinical response of human HGG to radiotherapy has not been neuronal markers), canine in origin according to the karyotypic demonstrated (45, 46), this parameter is widely used to compare the analysis and able to give rise to colonies according to the colony radiosensitivity of different cancer cell lines or to assess the forming efficiency results. Our phenotypic results are in agreement radiosensitizing potential of new therapeutics in vitro (47, 48). As Frontiers in Veterinary Science 06 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 and 4.84 Gy, respectively. However, SF2 values we obtained in this study are similar for J3T cell lines, higher for J3T-Bg and G06A compared to another recent study (49). These differences could be explained by the different culture medium between our two studies which could impact intrinsic radiosensitivity of cell lines. These values are higher than those described in the literature for human HGG cell lines (45). Indeed, the study of Taghian et al. evaluated SF2 and MID of 21 human HGG cell lines and reported a mean SF2 of 51% and a mean MID of 2.57 Gy. In this latter study, a wide variation in SF2 was observed between cell lines: some were obviously radioresistant (highest SF2 values between 70 and 80%) while others exhibited moderate radioresponsiveness (lowest SF2 values between 20 and 40%). Such a variation in SF2 has not been found in our study, maybe because of the limited number of cell lines of canine HGG. From various sarcomas and carcinomas, in the study of Maeda et al. for which the response of 27 canine cell lines to RT was evaluated, SF2 values as high as those of our study were described for mammary carcinoma and various sarcomas (50). The alpha and beta parameters provide information on the response of tumor cells to lethal and sublethal damages. Relevant information regarding the way to optimize RT treatment planning can also be provided by these parameters as the α/β ratio reflects the fractionation sensitivity of cells. For human HGG cell lines, the literature data indicate intermediate α/βvalues (5–10 Gy) (45, 51) and a mean alpha around 0.3 (52). By contrast with our study, the mean alpha/beta ratio and the mean alpha are, respectively, 0.836 Gy and 0.027 Gy−1. Such low values are only found in human prostastic cancers and malignant melanoma (53, 54). According to the quadratic model, tumors exhibiting an α/β ratio under 2 Gy are more sensitive to hypofractionation. Our results are in agreement with the results of a recent study showing that a hypofractionated radiation protocol performed in dogs for the treatment of intracranial tumors (10 × 4 Gy) leads to the same overall FIGURE 3 survival as the classical approach (20 × 2.5 Gy) without additional Pathologic features of xenografts: growth patterns (A–C), cell side effects. morphology and architectural patterns (D–H). (A) Focal infiltration, tumor G06A (H&E, scale bar: 500 μm): the interface between the The intracerebral implantation of the five canine cell lines tumor and the adjacent parenchyma, albeit readily identifiable, is available to us in the laboratory has confirmed their ability to multifocally infiltrated. (B) Diffuse infiltration, tumor J3T (H&E, scale reproduce high-grade gliomas of malignancy when they are bar: 500 μm): a diffuse neoplastic infiltrate is present within the caudate putamen (star) parietal cortex and leptomeninges (arrow). implanted orthotopically: astrocytomas for J3T, SDT3 and Raffray, cc, corpus callosum; lf, longitudinal fissure of the brain; LV, lateral oligodendroglioma for G06A and poorly differentiated glioma for ventricle. (C) Diffuse infiltration associated with perivascular cuffing, J3T-Bg. We confirmed that these cells lines were diffuse high-grade J3T-Bg tumor (H&E, scale bar: 500 μm): coalescence of large perivascular cuffs forms a multinodular and poorly delineated mass. glioma as they were all positive for Olig2 except for J3T-Bg cell (D) J3T-Bg tumor (H&E, scale bar: 50 μm): spindle cells organized in lines, and astrocytic type for those positive for GFAP and fascicles. (E) J3T tumor (H&E, scale bar: 100 μm): polygonal cells Vimentine (55). In addition of Olig2, CNPase marker could organized in packets separated by a fine capillary network. (F) Tumor J3T-Bg (H&E, scale bar: 50 μm): polygonal cells organized in be used with G06A cell lines to confirm the oligodendroglial perivascular cuffs. (G) J3T tumor (H&E, scale bar: 100 μm): differentiation of this cell lines (14, 56). The xenograft phenotype coexistence of a polygonal cell area and a spindle cell area separated is consistent with that of the original tumor for J3T cells (from from each other by the corpus callosum (arrow). (H) Tumor G06A (H&E, scale bar: 50 μm): sheets of round to polygonal cells with anaplastic astrocytoma) and SDT3 (from glioblastoma). The very hyperchromatic nucleus and a pale perinuclear halo, associated with little differentiated character of J3T-Bg can be explained by the fact nuclear rowing (arrows). (I) J3T tumor (H&E, scale bar: 100 μm): that these cells were recovered after subcutaneous implantation of numerous branching thin-walled capillaries within the tumor core. (J) J3T tumor (H&E, scale bar: 100 μm): geographic necrosis with J3T cells in Biege Nude XID mice, this passage potentially having palisading at the edges. led to the selection of a cell subpopulation (16). The Raffray cells were isolated from a grade III oligodendroglioma. In vitro analysis of the expression profile of the cell line carried out by the laboratory that isolated them revealed a poorly differentiated astrocytic it takes into account the entire survival curve, MID is also a useful phenotype, a phenotype confirmed by the analysis we carried out parameter for quantifying differences in survival curves between on these cells included in HistoGel. The orthotopic implantation cell lines. In our study, the mean SF2 and the mean MID were 84% of Raffray therefore confirms what had been observed in vitro: the Frontiers in Veterinary Science 07 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 FIGURE 4 Immunohistochemical characterization of xenografts. (A) Raffray tumor (IHC Olig2, scale bar: 50 μm): nuclear staining is intense in more than 90% of the tumor population (IHC score: ++++). (B) Raffray tumor, same tumor area as shown in panel (A) (IHC GFAP, scale bar: 50 μm): 60 to 90% of the cells exhibit a cytoplasmic staining of medium intensity (IHC score: ++). (C) SDT3 tumor (IHC GFAP, scale bar: 500 μm): in this focally infiltrative tumor, GFAP labeling can be seen at the interface between tumor and adjacent parenchyma, but also in the tumor core where 30 to 60% of the cells are positive with a moderately intense finely granular cytoplasmic labeling (IHC score: ++). Inset (scale bar: 50 μm): a labeled mitotic tumor cell is present in the center of the image. (D) G06A (IHC GFAP, scale bar: 500 μm): in this focally infiltrative tumor, GFAP labeling can only be seen at the interface between tumor and adjacent parenchyma and corresponds to immunostaining of reactive normal astrocytes. There is no labeling of cells within the tumor core (inset scale bar: 50 μm), few normal astrocyte processes interspersed between neoplastic cells are labeled (IHC score: 0). (E) Raffray tumor, same tumor area as in panels (A,B) (IHC Vimentin, scale bar: 50 μm): tumor cells are negative to Vimentin (IHC score: 0). (F) J3T-Bg (IHC Vimentin, scale bar: 50 μm): perivascular cuffs exhibit a weak cytoplasmic immunolabeling (IHC score: +). isolated tumor cells reproduce a high-grade astrocytoma and not make them the least interesting cell line to use in a perspective of an oligodendroglioma, which seems to indicate a selection of future implementation of a model of murine study of canine astrocytic tumor cells during the establishment of the cell line. diffuse glioma. Finally, the tumors obtained from the G06A do not reproduce the There is growing evidence that pet dogs with spontaneous HGG original tumor both in terms of phenotype (the xenograft is are relevant models for studying human diffuse gliomas (8, 10). These oligodendroglial, the tumor of astrocytic origin) and of grade of dogs therefore appear as potential animal models for preclinical malignancy (grade III for the xenograft, grade IV for the original testing of novel treatment strategies. Here, we achieved a tumor). This difference between the original tumor and xenograft radiobiological characterization of five canine HGG cell lines (J3T, for G06A and Raffray could be explained by intrinsic plasticity of J3T-Bg, SDT3, G06A and Raffray). Their general characteristics and glial tumors, leading to slight modifications in tumor phenotype their ability to generate tumors in immunocompromised mice were and markers expression between cell culture and xenograft (57). also evaluated. Among these five canine HGG cell lines, J3T, J3T-Bg That drift for G06A and Raffray from rate of engraftment of G06A and SDT3 validate the value of the canine glioma model for cells (only 25%) and the phenotype that they induce when injected further research. Frontiers in Veterinary Science 08 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 Data availability statement Conflict of interest The raw data supporting the conclusions of this article will The authors declare that the research was conducted in the be made available by the authors, without undue reservation. absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Ethics statement Publisher’s note The animal study was approved by Institution animal ethics committee with authorization number 2018031614514282. The study All claims expressed in this article are solely those of the authors was conducted in accordance with the local legislation and and do not necessarily represent those of their affiliated organizations, institutional requirements. or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Author contributions BC and AD: writing – original draft, conceptualization, and Supplementary material methodology. CD: conceptualization and resources. JA: resources. MP: conceptualization, resources, and supervision. EC-J and GM: The Supplementary material for this article can be found online supervision. All authors contributed to the article and approved the at: https://www.frontiersin.org/articles/10.3389/fvets.2023.1253074/ submitted version. full#supplementary-material References 1. Paoloni M, Khanna C. Translation of new cancer treatments from pet dogs to patient as a naturally occurring preclinical model for human Glioma. J Neuropathol Exp humans. Nat Rev Cancer. (2008) 8:147–56. doi: 10.1038/nrc2273 Neurol. (2018) 77:1039–54. doi: 10.1093/jnen/nly085 2. Snyder JM, Shofer FS, Winkle TJ, Massicotte C. Canine intracranial primary 15. Higgins RJ, Dickinson PJ, LeCouteur RA, Bollen AW, Wang H, Wang H, et al. neoplasia: 173 cases (1986–2003). J Vet Intern Med. (2006) 20:669. doi: Spontaneous canine gliomas: overexpression of EGFR, PDGFRalpha and IGFBP2 10.1111/j.1939-1676.2006.tb02913.x demonstrated by tissue microarray immunophenotyping. J Neuro-Oncol. (2010) 98:49–55. doi: 10.1007/s11060-009-0072-5 3. Song RB, Vite CH, Bradley CW, Cross JR. Postmortem evaluation of 435 cases of intracranial neoplasia in dogs and relationship of neoplasm with breed, age, and body 16. York D, Higgins RJ, LeCouteur RA, Wolfe AN, Grahn R, Olby N, et al. TP53 weight. J Vet Intern Med. (2013) 27:1143–52. doi: 10.1111/jvim.12136 mutations in canine brain tumors. Vet Pathol. (2012) 49:796–801. doi: 4. Kishimoto TE, Uchida K, Chambers JK, Kok MK, Son NV, Shiga T, et al. A 10.1177/0300985811424734 retrospective survey on canine intracranial tumors between 2007 and 2017. J Vet Med 17. Boudreau CE, York D, Higgins RJ, LeCouteur RA, Dickinson PJ. Molecular Sci. (2020) 82:77–83. doi: 10.1292/jvms.19-0486 signalling pathways in canine gliomas: glioma pathways. Vet Comp Oncol. (2015) 5. Miller AD, Miller CR, Rossmeisl JH. Canine primary intracranial cancer: a 15:133–50. doi: 10.1111/vco.12147 clinicopathologic and comparative review of Glioma, meningioma, and choroid plexus 18. Dickinson PJ, Sturges BK, Higgins RJ, Roberts BN, Leutenegger CM, Bollen AW, tumors. Front Oncol. (2019) 9:1151. doi: 10.3389/fonc.2019.01151 et al. Vascular endothelial growth factor mRNA expression and Peritumoral edema in 6. Lipsitz D, Higgins RJ, Kortz GD, Dickinson PJ, Bollen AW, Naydan DK, et al. canine primary central nervous system tumors. Vet Pathol. (2008) 45:131–9. doi: Glioblastoma multiforme: clinical findings, magnetic resonance imaging, and pathology 10.1354/vp.45-2-131 in five dogs. Vet Pathol. (2003) 40:659–69. doi: 10.1354/vp.40-6-659 19. Mitchell D, Chintala S, Fetcko K, Henriquez M, Tewari BN, Ahmed A, et al. 7. Magalhães TR, Benoît J, Néčová S, North S, Queiroga FL. Outcome after radiation Common molecular alterations in canine Oligodendroglioma and human malignant therapy in canine intracranial Meningiomas or Gliomas. In Vivo. (2021) 35:1117–23. gliomas and potential novel therapeutic targets. Front Oncol. (2019) 9:780. doi: 10.3389/ doi: 10.21873/invivo.12357 fonc.2019.00780 8. Dickinson PJ, York D, Higgins RJ, LeCouteur RA, Joshi N, Bannasch D. 20. Filley A, Henriquez M, Bhowmik T, Tewari BN, Rao X, Wan J, et al. Immunologic Chromosomal aberrations in canine Gliomas define candidate genes and common and gene expression profiles of spontaneous canine oligodendrogliomas. J Neuro-Oncol. pathways in dogs and humans. J Neuropathol Exp Neurol. (2016) 75:700–10. doi: (2018) 137:469–79. doi: 10.1007/s11060-018-2753-4 10.1093/jnen/nlw042 21. Spugnini EP, Thrall DE, Price GS, Sharp NJ, Munana K, Page RL. Primary 9. Amin SB, Anderson KJ, Boudreau CE, Martinez-Ledesma E, Kocakavuk E, Johnson irradiation of canine intracranial masses. Vet Radiol Ultrasound. (2000) 41:377–80. doi: KC, et al. Comparative molecular life history of spontaneous canine and human 10.1111/j.1740-8261.2000.tb02091.x Gliomas. Cancer Cell. (2020) 37:243–257.e7. doi: 10.1016/j.ccell.2020.01.004 22. Rossmeisl JH, Garcia PA, Daniel GB, Bourland JD, Debinski W, Dervisis N, et al. 10. Connolly NP, Shetty AC, Stokum JA, Hoeschele I, Siegel MB, Miller CR, et al. Cross- Invited review-Neuroimaging response assessment criteria for brain tumors in species transcriptional analysis reveals conserved and host-specific neoplastic processes veterinary patients: brain tumor response assessment. Vet Radiol Ultrasound. (2014) in mammalian glioma. Sci Rep. (2018) 8:1180. doi: 10.1038/s41598-018-19451-6 55:115–32. doi: 10.1111/vru.12118 11. Stoica G, Lungu G, Martini-Stoica H, Waghela S, Levine J, Smith R. Identification 23. Fertil B, Dertinger H, Courdi A, Malaise EP. Mean inactivation dose: a useful of cancer stem cells in dog glioblastoma. Vet Pathol. (2009) 46:391–406. doi: 10.1354/ concept for intercomparison of human cell survival curves. Radiat Res. (1984) 99:73–84. vp.08-VP-0218-S-FL doi: 10.2307/3576448 12. Fernández F, Deviers A, Dally C, Mogicato G, Delverdier M, Cauzinille L, et al. 24. Fertil B, Malaise EP. Intrinsic radiosensitivity of human cell lines is correlated Presence of neural progenitors in spontaneous canine gliomas: a histopathological and with radioresponsiveness of human tumors: analysis of 101 published survival immunohistochemical study of 20 cases. Vet J. (2016) 209:125–32. doi: 10.1016/j. curves. Int J Radiat Oncol Biol Phys. (1985) 11:1699–707. doi: tvjl.2015.10.039 10.1016/0360-3016(85)90223-8 13. Kishimoto TE, Uchida K, Thongtharb A, Shibato T, Chambers JK, Nibe K, et al. 25. Björk-Eriksson T, West C, Karlsson E, Mercke C. Tumor radiosensitivity (SF2) is Expression of Oligodendrocyte precursor cell markers in canine Oligodendrogliomas. a prognostic factor for local control in head and neck cancers. Int J Radiat Oncol Biol Vet Pathol. (2018) 55:634–44. doi: 10.1177/0300985818777794 Phys. (2000) 46:13–9. doi: 10.1016/S0360-3016(99)00373-9 14. Koehler JW, Miller AD, Miller CR, Porter B, Aldape K, Beck J, et al. A revised 26. Berens ME, Giese A, Shapiro JR, Coons SW. Allogeneic astrocytoma in immune diagnostic classification of canine Glioma: towards validation of the canine Glioma competent dogs. Neoplasia. (1999) 1:107–12. doi: 10.1038/sj.neo.7900020 Frontiers in Veterinary Science 09 frontiersin.org Cartiaux et al. 10.3389/fvets.2023.1253074 27. Rainov NG, Koch S, Sena-Esteves M, Berens ME. Characterization of a canine associated with cellular resistance to ionizing radiation in HeLa cells. Int J Cancer. (2002) glioma cell line as related to established experimental brain tumor models. J Neuropathol 100:662–8. doi: 10.1002/ijc.10558 Exp Neurol. (2000) 59:607–13. doi: 10.1093/jnen/59.7.607 42. McMahon SJ. The linear quadratic model: usage, interpretation and challenges. 28. Inoue S, Ichikawa T, Kurozumi K, Maruo T, Onishi M, Yoshida K, et al. Novel Phys Med Biol. (2018) 64:01TR01. doi: 10.1088/1361-6560/aaf26a animal glioma models that separately exhibit two different invasive and angiogenic 43. Deacon J, Peckham MJ, Steel GG. The radioresponsiveness of human tumours and phenotypes of human glioblastomas. World Neurosurg. (2012) 78:670–82. doi: 10.1016/j. the initial slope ofthe cell survival curve. Radiother Oncol. (1984) 2:317–23. doi: 10.1016/ wneu.2011.09.005 S0167-8140(84)80074-2 29. Chen Y-T, Tan KA, Pang LY, Argyle DJ. The class I PI3K/Akt pathway is critical for 44. Franken NAP, Rodermond HM, Stap J, Haveman J, van Bree C. Clonogenic assay cancer cell survival in dogs and offers an opportunity for therapeutic intervention. BMC of cells in vitro. Nat Protoc. (2006) 1:2315–9. doi: 10.1038/nprot.2006.339 Vet Res. (2012) 8:73. doi: 10.1186/1746-6148-8-73 45. Taghian A, Suit H, Pardo F, Gioioso D, Tomkinson K, DuBois W, et al. In vitro 30. Lai SR, Castello SA, Robinson AC, Koehler JW. In vitro anti-tubulin effects of intrinsic radiation sensitivity of glioblastoma multiforme. Int J Radiat Oncol. (1992) mebendazole and fenbendazole on canine glioma cells. Vet Comp Oncol. (2017) 23:55–62. doi: 10.1016/0360-3016(92)90543-Q 15:1445–54. doi: 10.1111/vco.12288 46. Taghian A, Ramsay J, Allalunis-Turner J, Budach W, Gioioso D, Pardo F, et al. 31. Schlein LJ, Fadl-Alla B, Pondenis HC, Lezmi S, Eberhart CG, LeBlanc AK, et al. Intrinsic radiation sensitivity may not be the major determinant of the poor clinical Immunohistochemical characterization of Procaspase-3 overexpression as a Druggable outcome of glioblastoma multiforme. Int J Radiat Oncol. (1993) 25:243–9. doi: target with PAC-1, a Procaspase-3 activator, in canine and human brain cancers. Front 10.1016/0360-3016(93)90345-V Oncol. (2019) 9:96. doi: 10.3389/fonc.2019.00096 47. Malric L, Monferran S, Delmas C, Arnauduc F, Dahan P, Boyrie S, et al. Inhibiting 32. Feldhaeusser B, Platt SR, Marrache S, Kolishetti N, Pathak RK, Montgomery DJ, integrin β8 to differentiate and Radiosensitize Glioblastoma-initiating cells. Mol Cancer et al. Evaluation of nanoparticle delivered cisplatin in beagles. Nanoscale. (2015) Res. (2019) 17:384–97. doi: 10.1158/1541-7786.MCR-18-0386 7:13822–30. doi: 10.1039/C5NR03447G 48. Lacore MG, Delmas C, Nicaise Y, Kowalski-Chauvel A, Cohen-Jonathan-Moyal E, 33. Debinski W, Dickinson P, Rossmeisl JH, Robertson J, Gibo DM. New agents for Seva C. The glycoprotein M6a is associated with invasiveness and radioresistance of targeting of IL-13RA2 expressed in primary human and canine brain tumors. PLoS One. Glioblastoma stem cells. Cells. (2022) 11:2128. doi: 10.3390/cells11142128 (2013) 8:e77719. doi: 10.1371/journal.pone.0077719 49. Schrock MS, Zalenski AA, Tallman MM, Kollin L, Bratasz A, Weeks G, et al. 34. York D, Sproul CD, Chikere N, Dickinson PJ, Angelastro JM. Expression and Establishment and characterization of two novel patient-derived lines from canine high- targeting of transcription factor ATF5 in dog gliomas. Vet Comp Oncol. (2018) 16:102–7. grade glioma. Vet Comp Oncol. (2023) 21:492–502. doi: 10.1111/vco.12912 doi: 10.1111/vco.12317 50. Maeda J, Froning CE, Brents CA, Rose BJ, Thamm DH, Kato TA. Intrinsic 35. Koehler J, Sandey M, Prasad N, Levy SA, Wang X, Wang X. Differential expression radiosensitivity and cellular characterization of 27 canine cancer cell lines. PLoS One. of miRNAs in hypoxia (“HypoxamiRs”) in three canine high-grade glioma cell lines. (2016) 11:e0156689. doi: 10.1371/journal.pone.0156689 Front Vet Sci. (2020) 7:104. doi: 10.3389/fvets.2020.00104 51. Barazzuol L, Burnet NG, Jena R, Jones B, Jefferies SJ, Kirkby NF. A mathematical 36. Dupont-Monod J., Abadie J.Université Claude Bernard (Lyon), et VetAgro Sup, model of brain tumour response to radiotherapy and chemotherapy considering (2014). « Etablissement et exploitation d’une lignée cellulaire de gliome canin pour la radiobiological aspects. J Theor Biol. (2010) 262:553–65. doi: 10.1016/j.jtbi.2009.10.021 recherche thérapeutique », [sn], Sl. 52. Schultz CT, Geard CR. Radioresponse of human astrocytic tumors across grade as 37. Berens ME, Bjotvedt G, Levesque DC, Rief MD, Shapiro JR, Coons SW. a function of acute and chronic irradiation. Int J Radiat Oncol. (1990) 19:1397–403. doi: Tumorigenic, invasive, karyotypic, and immunocytochemical characteristics of clonal 10.1016/0360-3016(90)90350-S cell lines derived from a spontaneous canine anaplastic astrocytoma. In Vitro Cell Dev Biol Anim. (1993) 29:310–8. doi: 10.1007/BF02633959 53. Hennequin C, Dubray B. Le rapport alpha/bêta revisité à l’heure de l’hypofractionnement. Cancer Radiothér. (2013) 17:344–8. doi: 10.1016/j. 38. Dahan P, Martinez Gala J, Delmas C, Monferran S, Malric L, Zentkowski D, et al. canrad.2013.06.035 Ionizing radiations sustain glioblastoma cell dedifferentiation to a stem-like phenotype through survivin: possible involvement in radioresistance. Cell Death Dis. (2014) 54. Wang JZ. How low is the alpha/beta ratio for prostate cancer? Int J Radiat Oncol 5:e1543. doi: 10.1038/cddis.2014.509 Biol Phys. (2003) 55:1–15. doi: 10.1159/000071805 39. Joiner KS, Spangler EA. Evaluation of HistoGel™-embedded specimens for use in 55. Kosty J, Lu F, Kupp R, Mehta S, Lu QR. Harnessing OLIG2 function in veterinary diagnostic pathology. J Vet Diagn Invest. (2012) 24:710–5. doi: tumorigenicity and plasticity to target malignant gliomas. Cell Cycle. (2017) 16:1654–60. 10.1177/1040638712445763 doi: 10.1080/15384101.2017.1361062 40. Rebetz J, Tian D, Persson A, Widegren B, Salford LG, Englund E, et al. Glial 56. Johnson GC, Coates JR, Wininger F. Diagnostic immunohistochemistry of canine progenitor-like phenotype in low-grade glioma and enhanced CD133-expression and and feline Intracalvarial tumors in the age of brain biopsies. Vet Pathol. (2014) neuronal lineage differentiation potential in high-grade glioma. PLoS One. (2008) 51:146–60. doi: 10.1177/0300985813509387 3:e1936. doi: 10.1371/journal.pone.0001936 57. Yabo YA, Niclou SP, Golebiewska A. Cancer cell heterogeneity and plasticity: a 41. Dalenc F, Drouet J, Ader I, Delmas C, Rochaix P, Favre G, et al. Increased paradigm shift in glioblastoma. Neuro Oncol. (2022) 24:669–82. doi: 10.1093/neuonc/ expression of a COOH-truncated nucleophosmin resulting from alternative splicing is noab269 Frontiers in Veterinary Science 10 frontiersin.org