Establishment of TP53-knockout canine cells using optimized CRIPSR/Cas9 vector system for canine cancer research
Genetic engineering technology such as clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system provides a powerful tool for developing disease models and determining gene functions. Recent interests in canine cancer models have highlighted the necessity of developing genetic engineering tools for dogs. In this study, we attempted to generate optimized CRISPR/Cas9 system to target canine tumor protein 53 (TP53), one of the most crucial tumor suppressor genes, to establish TP53 knockout canine cells for canine cancer research.
We constructed CRISPR/Cas9 vectors using each of three TP53 gene-targeting guide RNAs (gRNAs) with minimal off-target potential. After transfection, we obtained several clones of TP53 knockout cells containing “indel” mutations in the targeted locus which had infinite cellular life span, resistance to genotoxicity, and unstable genomic status in contrast to normal cells. Of the established TP53 knockout cells, TP53KO#30 cells targeted by TP53 gRNA #30 showed non-cancerous phenotypes without oncogenic activation both in vitro and in vivo. More importantly, no off-target alteration was detected in TP53KO#30 cells. We also tested the developmental capacity of TP53 knockout cells after application of the somatic cell nuclear transfer technique.
Our results indicated that TP53 in canine cells was effectively and specifically targeted by our CRISPR/Cas9 system. Thus, we suggest our CRISPR/Cas9-derived canine TP53 knockout cells as a useful platform to reveal novel oncogenic functions and effects of developing anti-cancer therapeutics.
KeywordsCRISPR/Cas9 Dog TP53 Knockout Cancer
Canine cancer is among the most common causes of death of pet dogs. According to a report from the Swiss Canine Cancer Registry, 51.83% (67,943 of 126,693) of dogs were diagnosed with various cancers from 1955 to 2008 . Although canine cancer therapy has considerably improved and multiple genes were identified as candidate biomarkers to detect and cure the cancer, the prognosis of various canine cancer patients remains poor . Understanding molecular aspects of canine oncology will improve clinical outcomes. Canine cell models susceptible to tumorigenesis are the most essential materials to elucidate the molecular backgrounds underlying carcinogenesis and to find novel therapeutic targets. Moreover, recent interests in canine cancer models for use in the translational research field have also highlighted the necessity of developing research tools for dogs .
Tumor protein 53 (TP53) gene is a key regulator in cell cycle arrest, cellular senescence, and apoptotic responses induced by various stresses . TP53 is also known as the most crucial tumor suppressor gene and its mutation frequency was over one-third of pan-cancer patients [5, 6]. So, its importance in cancer initiation and progression, and in therapeutics has been well identified by numerous studies . Like in human cancer, genetic alteration in TP53 gene was frequently observed in various canine cancer including lymphoma and mammary cancer [8, 9]. So, canine TP53 modulating tools and canine experimental model of TP53 deficiency are the most fundamental requirement to study canine cancers.
Recently, the type II clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system, an RNA-guided nuclease-mediated adaptive immune system of Streptococcus pyogenes against phages and viruses, was reconstituted in eukaryotic cells via codon optimization and the unification of two CRISPR RNA components, the guide RNA (gRNA) and trans-activating CRISPR RNA, into a single guide RNA [10, 11, 12, 13]. Double strand breaks (DSBs) generated by its two nuclease domains, HNH and RuvC, are then restored via one of two cellular repair systems: non-homologous end-joining and homology-directed repair pathways. The former produces a random insertion or deletion (indel) mutation around the DSB site, while the latter introduces precise insertion of an intended DNA sequence from a designed donor template . However, the potential off-target activity of the RNA-guided CRISPR/Cas9 system causing unintended genetic alterations is a major concern in basic and clinical applications. . Therefore, minimizing the off-target potential of this system is critical for obtaining precise results.
In this study, we constructed a CRISPR/Cas9 vector system for canine TP53 with minimum off-target potential and TP53 knockout canine fibroblasts using the system, and finally evaluate their utilities in cancer studies.
Construction of CRISPR/Cas9 systems for canine TP53 gene knockout
TP53 KO canine fibroblasts were immortalized and acquired resistance against genotoxic stress
The TP53 protein expressed by TP53 is engaged in cell cycle arrest and DNA repair mechanisms after DSBs; therefore, its deficiency is closely related to resistance to ionizing radiation or genotoxic chemicals [16, 17]. During this process, the cyclin-dependent kinase inhibitor p21 is transcriptionally upregulated by activated TP53 protein and functions in DNA damage-induced cell arrest . Thus, we treated the TP53KO#30 and TP53KO#39 cell lines with temozolomide (TMZ), a DNA alkylating agent, to examine whether TP53 function was intact. As determined by the levels of p21 protein in each cell after TMZ treatment for 48 h, in sharp contrast to K9 fetus 1 control cells, p21 protein was not apparently induced by TMZ in TP53 KO cells and cells in which TP53 was inactivated by SV40LT (Fig. 3g). These results indicate that both CRISPR/Cas9-mediated TP53 KO fibroblast lines are resistant to DNA damage, which is a typical characteristic of TP53-deficient cells.
Tumorigenic potential testing of CRISPR/Cas9-mediated TP53 KO canine cells
Developmental capacity of TP53KO#30 cells after somatic cell nuclear transfer (SCNT)
Pregnancy outputs from cloned embryos according to cell line
No. of oocyte donors dogs
No. of surrogate
No. of pregnancies (%)a
Oocyte subjected to NT
Oocyte fused & transferred (%)c
On Day 30
K9 Fetus 1 (control)
Despite advances in veterinary science, cancer remains a leading cause of death of pet dogs. Combining clinical studies and appropriate mechanistic studies of canine cancers will promote the development of advanced drugs and improve clinical outcomes. Recently, spontaneous canine cancer models have attracted the attention of the translational research field because of the various similarities in oncogenetic characteristics, histological features, and response to conventional therapies in humans . However, few research materials such as canine cell lines, genetic engineering tools, antibodies, and even bioinformatics database for canine models. Thus, various experimental tools for canine cancer research are urgently needed.
In this study, we designed CRISPR/Cas9 vector systems for canine TP53, an important tumor suppressor gene. CRISPR/Cas9 of TP53 gRNA #30 and #39, but not #51, effectively targeted canine TP53. We also examined the proliferation capacities, morphological changes, and cellular transformation of canine fetal fibroblast cells targeted by TP53 gRNA #30 and #39. In contrast to TP53KO#30 cells, TP53KO#39 cells showed abnormal phenotypes. As observed in an early TP53-knockout mouse model, depletion of TP53 alone is not sufficient to induce transformation into cancerous cells . Thus, abnormal characteristics of TP53KO#39 cells may be caused by undesired genetic alterations generated from off-target effects of the CRISPR/Cas9 or attenuation of genetic stability governed by TP53. One possibility is that a mutation in TFE3, a possible off-target gene for TP53 gRNA #39, caused cellular transformation. Therefore, our results suggest that the sequence of TP53 gRNA #30, which has a lower potential for off-target effects, is an effective tool for targeting canine TP53, and TP53KO#30 cells are TP53-targeted canine cells without off-target gene alterations. This first successful application of the CRISPR/Cas9 system in canine models may promote the development of various genetic engineering tools for a range of purposes in canine models.
TP53 deficiency is related to high incidences of human malignancies including lymphomas, soft-tissue carcinomas, or even Li-Fraumeni syndrome; therefore, TP53 KO mice have been used as a genetic background model for specific cancers, including glioma, ovarian cancer, medulloblastoma, and lung cancer by modulating genes of interest such as KRAS, NF1, or MYC [27, 28, 29, 30]. These cancer models have provided critical information of regarding oncogenic signaling during tumor initiation and progression. Additionally, many reports have utilized cancer cell models, generated by modulating TP53 and/or oncogenes, to evaluate the efficacy of developing drug [31, 32]. Genetic engineering tools targeting TP53 and TP53-deficient cell lines are among the most fundamental experiment materials useful for various cancer studies, particularly for studying the pathological functions of unidentified oncogenes in canine cancers and developing therapeutics.
TP53 KO alone does not cause any critical problems in early mouse development ; therefore, if TP53KO#30 cells exhibit normal characteristics, it should be possible to generate TP53 KO dogs via SCNT. Thus, we attempted to generate TP53 KO dogs via SCNT using TP53KO#30 cells as nuclear donor cells. However, the SCNT embryos derived from TP53KO#30 cells failed to initiate pregnancy in surrogate dogs. These failures in embryo development and pregnancy may have been caused by chromosome abnormalities in TP53KO#30 cells [33, 34]. Chromosomal missegregation did not occur frequently during early embryogenesis of TP53 KO mice; however, a tetraploid cell population was increased in postnatal TP53 KO mice . TP53 loss in normal human fibroblasts and even epithelial cancer cells does not always result in aneuploidy . These previous studies showed that disruption of TP53-governing genetic defense mechanisms leads to chromosomal abnormality or genetic alterations by chance during constant proliferation. Thus, immediate SCNT after transfection of our CRISPR/Cas9 system may be useful for producing TP53 KO dogs to avoid the occurrence of chromosomal abnormalities. Additionally, it may be more scientifically valuable to produce transgenic canine models of specific cancers by combining our canine TP53 targeting system with tissue or cell type-specific expression systems. Therefore, our TP53-targeting system may be utilized to produce TP53-modulated canine transgenic models in further studies.
We constructed an optimal CRISPR/Cas9-mediated TP53 targeting system for dogs with minimal off-target potential and established TP53 KO canine cells, TP53KO#30, by applying our system. Because of the weak off-target activity, the established TP53KO#30 cells showed typical characteristics of TP53-deficient cells without any cancerous properties. Therefore, our CRISPR/Cas9-mediated TP53 targeting system and TP53KO#30 cells can be utilized as a useful experimental materials for producing in vitro and in vivo canine cancer models to study the biology of canine.
Canis lupus familiaris canine fetal fibroblast cells (K9 Fetus 1), were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Life Technologies), 1% Glutamax (Life Technologies), 1% minimum essential medium non-essential amino acids (Life Technologies), 1% antibiotic-antimycotic (Life Technologies), and 0.1% 2-mercaptoethanol (Life Technologies).
For the temozolomide (TMZ; Sigma Aldrich, St. Louis, MO, USA) treatment experiment, cells were harvested using Trypsin-EDTA, counted, and reseeded at a density of 1 × 106 cells per 100-mm culture plate. Twelve hours later, 100 nM of TMZ was added to each well.
Guide RNA design
To design gRNAs targeting the canine TP53 locus, all candidates with NGG-protospacer adjacent motif (PAM) sequences in all exons of the gene were identified using Feng Zhang laboratory’s Target finder program (https://zlab.bio/guide-design-resources). The program’s automatic off-target screening system was not available for dog species; therefore, we narrowed the results using the following steps. First, gRNAs near the N-terminus within the third exon of each gene were selected because their partial functional products could be produced if their functional domains were involved. Next, candidates with a lower binding potential compared to other genomic regions (gRNAs ≤ max score 30.2), particularly to coding sequences, were identified using the Basic Local Alignment Search Tool (BLAST) to avoid off-target alterations. Nineteen gRNAs for TP53 were selected using this process and the final three gRNAs were selected based on their lowest non-specific binding potential, in accordance with the Max score determined using the BLAST algorithm.
To construct a CRISPR/Cas9 plasmid vector targeting canine TP53, the U6-stuffer-hSpCas9 sequence was digested with Kpn1 (blunted by Klenow fragment) and BamH1 from lentiCRISPRv2 (#52961; Addgene, Cambridge, MA, USA) . The fragment was inserted into the Nru1 and BamH1 sites of pcDNA3.1(+) (Thermo Fisher Scientific, Waltham, MA, USA). The enhanced green fluorescence protein (EGFP) sequence was digested with EcoR1 and Not1 from pEGFP-N2 (Clontech Laboratories Inc., Mountain View, CA, USA) and the synthesized E2A peptide sequence was inserted into the pcDNA3.1-CRISPR plasmid. Finally, the three designed gRNA sequences were synthesized and inserted into the two BsmB1 sites flanking the stuffer sequence.
Construction of canine TP53 KO fibroblasts
CRISPR/Cas9 vectors targeting TP53 were transfected transiently using Polyexpress™ (Excellgene, Rockville, MD, USA) into K9 fetus 1 cells at passage 1 according to the manufacturer’s instructions. The cells were then cultured and transferred into culture plates until untransfected cells showed senescence phenotypes. Finally, cell colonies with a normal cellular morphology and extended life span were separated and maintained independently.
Semi-solid agar culture assay
A 1.5 mL lower layer of 0.72% agar in DMEM containing 10% FBS was added to each well of a 6-well plate and allowed to solidify at room temperature (19–21 °C). Cells tested by the semi-solid agar assay were suspended in a plating layer (1.5 mL) of 0.28% agar in DMEM containing 10% FBS and added to the wells. The agar was allowed to solidify at room temperature and then incubated at 37 °C in a humidified CO2 incubator. To prevent the agar from drying, 500 μL of DMEM containing 10% FBS was added to each well every 2–3 days.
Cell growth counting
For cell proliferation analysis, 1.5 × 104 K9 fetus 1 control cells and TP53 KO cells were seeded into 6-well plates and cultured in the presence of 10% FBS for 6 days. Cells were harvested via trypsinization at various times and stained with Trypan blue. Viable cells were counted using a hematocytometer and inverted microscope. The number of cells was averaged over three independent experiments.
For western blot analysis, whole cell extracts were prepared in radioimmunoprecipitation assay lysis buffer (150 mM NaCl, 1% NP-40, 0.1% SDS, and 50 mM Tris, pH 7.4) containing 1 mM β-glycerophosphate, 2.5 mM sodium pyrophosphate, 1 mM NaF, 1 mM Na3VO4, and a protease inhibitor cocktail (Roche, Basel, Switzerland). Protein levels were quantified using the Bradford assay reagent (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s instructions. Proteins were separated by SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Pall Corporation, Port Washington, NY, USA) according to standard protocols. Membranes were immunoblotted with antibodies against HRAS (Merck Millipore, Billerica, MA, USA), TP53, p21, SV40 T antigen, and β-Actin (Santa Cruz Biotechnology, Dallas, TX, USA) in 3% bovine serum albumin in TBST. After primary antibody incubation, the membranes were washed and probed with horseradish peroxidase-conjugated goat anti-mouse or anti-rabbit IgG secondary antibodies (Pierce Biotechnology, Rockford, IL, USA). β-Actin was used as a loading control.
Canine TP53 target sequencing
Genomic DNA was extracted from canine fetal fibroblasts by using a Wizard Genomic DNA Extraction Kit (Promega, Madison, WI, USA), according to the manufacturer’s instructions. Next, exon 3 of TP53 in each cell line was amplified using Takara Ex taq™ (Takara Bio Inc., Shiga, Japan) and primer sets, forward 5′-CTGGTAAGGACTGGGTGTGG-3′ and reverse 5′-GCCACTGACCGTCCAAGTAA-3′under the following conditions in the thermal cycler: 60 s at 95 °C; 30 cycles of: 30 s at 95 °C, 40 s at 58 °C, and 60 s at 72 °C; followed by 10 min at 72 °C. Each product was ligated into a pGEM T-easy vector (Promega), according to the manufacturer’s instructions, and then sequenced using an ABI BigDye® terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA).
In vivo tumorigenesis assay
To establish subcutaneous xenograft models, 2 × 106 cells of each cell line in phosphate-buffered saline were mixed with 50% Matrigel (Invitrogen, Carlsbad, CA, USA) and transplanted subcutaneously into 5–6-week-old BALB/c nu/nu mice. The size of each tumor was calculated using the following formula: tumor volume (mm3) = longest diameter of tumor (mm) × shortest diameter of tumor (mm)2/2.
Normal canine and TP53 KO fibroblasts were karyotyped commercially by GenDix, Inc. (Seoul, Korea). The results of karyotyping of the two fibroblast lines were analyzed using the ChIPS-Karyo program, a chromosome image processing system. Analysis data and all terms were represented by an International System for Human Cytogenomic Nomenclature 2016 (ISCN 2016).
Potential off-target genes of each gRNA were predicted using the BLAST algorithm and final target genes for the surveyor assay were selected in accordance with the matched number of nucleotides to each TP53 gRNA (≥10 of 20) and presence of a NGG PAM sequence.
Genomic DNA was extracted from canine fetal fibroblasts and TP53 KO cells using a Wizard Genomic DNA Extraction Kit (Promega) according to the manufacturer’s instructions. The surveyor assay was conducted using the Guide-it™ Mutation Detection Kit (Clontech Laboratories, Inc.) according to the manufacturer’s instructions. During the amplification step, PCR was performed using 100 ng of extracted canine genomic DNA. Primer sequences are listed in Additional file: Table S2.
Unless otherwise indicated, all reagents were obtained from Sigma-Aldrich. All donors and recipients employed in the study showed spontaneous estrous. The estrous stage was examined weekly by observing for vulval bleeding to detect the onset of the heat period. During heat, a 2 mL blood sample was collected daily by cephalic venipuncture and serum P4 levels in the blood samples were measured by electrochemiluminescence immunoassay (Cobas e411, Roche Diagnostics, Mannheim, Germany; intra- and inter-assay coefficients of variation < 4%). Ovarian ultrasonographies were periodically performed twice per day when serum P4 levels were found to be increased by more than 2 ng/mL. The time of ovulation was designated as the time when the ovaries became difficult to find for an apparent decrease in the number or contour of anechoic follicles, or for their disappearance anechogenicity by transabdominal ultrasonography and as the proportion of cornified cells was greater than or equal to 90% of epithelial cells from vaginal swabs, which were stained following Diff Quik (Sysmex Co., Kobe, Japan) standard protocols .
All oocyte donors and surrogates underwent spontaneous estrous, and donors and surrogates were matched based on the synchronization of their estrus. Oocytes were surgically retrieved at 3–4 days post-ovulation. Before surgery, a blood sample was drawn through the cephalic venipuncture, and blood plasma was collected and frozen (− 20 °C) for hormone analyses. Anesthesia was induced with a mixture of xylazine hydrochloride (Rumpun®; Bayer Korea, Ansan, Korea; 1 mg/kg body weight) and ketamine HCl (Ketalar®; Yuhan Corporation; 50 mg/mL, Seoul, Korea; 4 mg/kg body weight) and maintained with isoflurane inhalational. Under aseptic conditions, the reproductive tract was exposed through a midventral incision. Corpora lutea (CL) were counted and oocytes were bilaterally flushed from each oviduct with 10 mL TCM 199 supplemented with HEPES (Invitrogen). Oocytes were collected using a stereomicroscope, transferred into fresh medium, and subjected to nuclear transfer.
Evaluation of retrieved oocytes
The maturation stage of the retrieved oocytes was determined as previously described . The oocytes were stripped of cumulus cells and pre-stained with 5 mg/mL Bisbenzimide (Hoechst 33342) to visualize the presence of nuclei for enucleation process. Oocytes were graded based on their morphology and nuclear stage as immature (cumulus very closely attached to oocytes, nuclear stage is either germinal vesicle (GV), GV breakdown, or metaphase I), mature (M II oocytes with several layers of cumulus cells and homogeneous cytoplasm), aged (unidentified nuclear status with the cytoplasmic membrane shrink, MII oocytes in less than 70% of the cytoplasm and loosely attached cumulus cells,), abnormal (irregular cytoplasmic contour, protrusion of zona pellucida, nuclear immaturity), or ruptured (oocytes with broken zona and cytoplasmic membrane) under an inverted microscope equipped with epiflurescence (TE2000-E; Nikon Corp., Tokyo, Japan).
Preparation of donor cells
Donor cells originated form dogs active in police or military service. Dermal tissue samples from 2 male Belgian Malinois breeds, and 1 male German, measuring approximately 1 × 3 cm were collected under light tranquilization (Zoletil 50® Virbac, Carros, France) at 0.1 mg/kg and local anesthesia (Daehan lidocaine HCl 2%, Dai Han Pharm Co., Ltd., Seoul, Korea). Sections of subcutaneous tissues were cut into small pieces (approximately 1 × mm2) and cultured in DMEM containing 10% FBS at 37 °C in an atmosphere of 5% CO2 and air to obtain fibroblasts. Explants were maintained in the culture until they approached 90% confluence. Cells were then trypsinized and reconstituted at concentrations of approximately 1 × 106 cells per mL, and then cryopreserved in cryovials containing DMEM containing 10% dimethyl sulfoxide.
After evaluating of the maturation status, metaphase II oocytes were enucleated by squeezing out the first polar body and metaphase II plate into a small amount of surrounding cytoplasm using a glass pipette. Donor cells, TP53KO#30 fibroblasts (passage 4), were prepared and treated using a conventional system of primary cell culture as described previously . Using a fine pipette, a trypsinized cells with a smooth cell surface was transferred into the perivitelline space of an enucleated oocyte. The couplets were equilibrated with 0.26 M mannitol solution containing 0.5 mM of HEPES, 0.1 mM of CaCl2, and MgSO4 for 4 min. Next, the couplets were transferred to a chamber with two electrodes and covered with mannitol solution. The couplets were fused with two DC pulses of 1.75–1.85 kV/cm for 15 μs using a BTX Electro-Cell Manipulator 2001 (BTX, Inc., San Diego, CA, USA). After simultaneous fusion and activation, a group of 5–6 embryos were cultured in 25 μL microdrops of mSOF covered with mineral oil for 1 h at 39 °C in a humidified atmosphere (5% O2, 5% CO2, and 90% N2) until embryo transfer.
Embryo transfer and pregnancy diagnosis
Surrogate dogs with estrus matching that of oocyte donors were anaesthetized as described previously using an oocyte retrieval procedure. The ovary with a greater number of corpus lutuea was approached by ventral laparotomy. The fat layer covering the ovary was gently grasped with forceps and suspended with a suture to exteriorize the fimbriated end of the oviduct. Immediately after fusion and activation, all reconstructed embryos were loaded into a tomcat catheter (3.5 Fr × 5.5″; Sherwood Medical, St. Louis, MO, USA) with at least a medium volume (2-4 μL) and gently transferred into the 2/3 distal position of the oviduct through the infundibulum. Pregnancy was confirmed by transabdominal ultrasound with a real-time ultrasonography at 25–30 days after embryo transfer. Ultrasonography was performed either in the standing or dorsal recumbency position using a portable ultrasound machine with a 3.5 MHz curved transducer (Sonace R7; Samsung Medison, Seoul, Korea). Ultrasonographies were repeated every 7 days on pregnant surrogates until term. The sizes and shapes of the chorionic cavities and presence of an embryonic or fetal heartbeat were examined to identify embryonic or fetal death.
All experiments were replicated more than three times. All data were analyzed by one-way ANOVA (analysis of variance) followed by Duncan’s test using SPSS software (SPSS, Inc., Chicago, IL, USA) and are reported as the mean ± standard error of the mean. Differences were considered significant if the P-value was less than 0.05.cancers and even for testing new anti-cancer therapeutics for dogs.
We are grateful to all members of the Cancer Growth Regulation Laboratory for their helpful discussion and technical assistance.
This work was supported by grants from the Next-Generation BioGreen21 Program (PJ01107701), Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries(IPET) through Agri-Bio industry Technology Development Program funded by Ministry of Agriculture, Food and Rural Affairs (Grant number: 318016-5), and the National Research Foundation (NRF) funded by the Ministry of Science (NRF-2016R1D1A1B03931941).
Availability of data and materials
All data generated or analyzed during this study are included in this published article and its Additional files.
KE and MGP conducted the experiments. MGP, YWJ, and YIJ contributed to the experimental procedures. KE and YWJ conducted data analysis. KE, SHK, and HK wrote the manuscript. SHH and WSH provided valuable opinions and assisted technically during nuclear transfer procedures. SHK and HK supervised all procedures. All authors read and approved the final manuscript.
All mouse experiments were approved by the animal care committee at the Korea University (Permit Number: KUIACUC-20161017-2) and carried out in accordance with government and institutional guidelines and regulations. All experimental procedures for dogs were conducted in accordance with the animal study guidelines and approved by the committee at the Sooam Biotech Research Foundation, Korea (permit no. C-12-01).
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 1.Grüntzig K, Graf R, Boo G, Guscetti F, Hässig M, Axhausen KW, Fabrikant S, Welle M, Meier D, Folkers G, et al. Swiss canine Cancer registry 1955–2008: occurrence of the Most common tumour diagnoses and influence of age, breed, body size, sex and neutering status on tumour development. J Comp Pathol. 2016;155(2):156–70.CrossRefGoogle Scholar
- 38.Johnston SD, Kustritz MVR, Olson PNS. Canine and feline theriogenology. Philadelphia: Saunders; 2001.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.