Abstract
Background/Aim: N1S1 rat models are commonly used in human medicine to study hepatocellular carcinoma (HCC). However, their use in veterinary medicine has not been reported. Thus, the aim of this study was to investigate whether the N1S1 rat models could be used to study canine HCC. Materials and Methods: The animals were divided into four groups: normal rat, N1S1 rat, normal dog, and HCC dog. Liver tissues of all animals were evaluated for vascular endothelial growth factor (VEGF), epidermal growth factor receptor (EGFR), platelet-derived growth factor receptor (PDGFR)-α, PDGFR-β, and c-kit by immunohistochemistry. Slides of each factor were scored according to the percentage of stained tumor cells and intensity of the staining. Results: Scores of VEGF and c-kit were high both in the tumor groups (the N1S1 rat and HCC dog groups) and the normal groups of dogs and rats. PDGFR-α was lower in the N1S1 rat group than that in the normal rat group (p=0.0042). It was also lower in the HCC dog group compared to the normal dog group (p=0.0008). PDGFR-β was higher in the HCC dog group than that in the normal dog group (p=0.0023) but was not detectable in the rat groups. EGFR was not detectable in any group. Conclusion: Based on immunochemistry results, PDGFR-α and PDGFR-β can be used as biomarkers of canine HCC. Because PDGFR-α showed consistency between rats and dogs, it can be used for studying canine HCC.
Hepatocellular carcinoma (HCC) is the most common primary neoplasm of the liver in dogs (1, 2) and the sixth most commonly occurring cancer in humans (3). Certain chemical agents can cause hepatic neoplasia experimentally (4). However, the cause of naturally occurring hepatic neoplasia in dogs is unknown (4). In human, hepatitis C virus, hepatitis B virus, heavy alcohol drinking, diabetes, and nonalcoholic fatty liver disease can cause HCC (5). HCC is classified as massive, nodular, and diffuse in dogs (6). The massive form is the most common one, followed by the nodular and diffuse forms (6, 7).
In humans, ablation is used to treat patients with very early stages of cancer, such as Barcelona Clinic Liver Cancer 0 (5, 8). Surgery, transplantation, and radioembolization are also included in the treatment (5, 8). Chemotherapy is one of the best options for patients when it is difficult to apply the other treatment options in humans (9). In veterinary medicine, although there are some reports of using gemcitabine or sorafenib for dogs with HCC, information on chemotherapy of HCC is insufficient (2, 10).
The prognosis of HCC in dogs is associated with outcomes of surgical resection, involved liver size, and levels of liver enzymes in blood chemistry (1). For the treatment of a massive form, surgery, such as lobectomy, is one of the options (6, 7). Nodular and diffuse forms have higher metastatic rates than the massive form (2). Therefore, the prognosis of a nodular or diffuse form is poor (2).
Tyrosine kinase receptors (TKRs) are cell-surface receptors that can activate signaling through tyrosine phosphorylation (11). They are involved in the regulation of growth and differentiation of normal cells (12). They are also key players in the regulation of oncogenesis (11). Thus, tyrosine kinase inhibitors (TKIs) are used as targeted treatments of tumors (12). TKIs can disturb the binding of TKs and cellular ATP so that downstream signaling is inhibited (12, 13). TKIs, such as toceranib, masitinib, and imatinib, are used for treating several tumors, such as mast cell tumors, osteosarcoma, and mammary gland tumors (13-17).
For studies of HCC, N1S1 or McA-RH7777 orthotopically implanted rat HCC models are used (18, 19). Through these models, complicated experiments can be conducted easily (18). Technical methods of efficacy or safety of anticancer drugs and assessment about the prognosis of HCC are advanced by using these models (3).
The aim of the present study was to determine whether N1S1 rat model could be used to study canine HCC. Rat HCC models inoculated with N1S1 cells were compared with the HCC dog using immunohistochemical staining of ligands and receptors associated with tyrosine kinases (TKs).
Materials and Methods
Study design and animals. In this study, four rats and two dogs were used. These four rats were all male Sprague Dawley rats (weight range=250-300 g). They were divided into two groups: a normal rat group (n=1) and an N1S1 rat group (n=3). N1S1 tumor cells were inoculated subcapsularly to the left lateral lobe of the liver of each rat in the N1S1 rat group. The shape and size of inoculated tumors were monitored by ultrasound every week for three weeks.
For a normal dog group (n=1), a healthy male Beagle dog was used. The beagle was managed and used under the guidelines of the Institutional Animal Care and Use Committee at Chonnam National University. For an HCC dog group (n=1), a dog that visited the Veterinary Teaching Hospital, Department of Veterinary Internal Medicine, Chonnam National University during 2021 was included. The owner agreed to provide the dog’s tumor tissue for this study. The dog was a 9-year-old, neutered male, Maltese. The mass of the dog was very large. It was identified across the right lobe of the liver through computed tomography (CT). The mass was diagnosed as HCC through histopathology.
The consistency in the results of immunohistochemistry (IHC) regarding the ligands and TKRs between each group was investigated. This study complied with the Institutional Animal Care and Use Committee guidelines at Chonnam National University (identification code No. CNU IACUC-YB-2021-112, No. CNU IACUC-YB-2021-166).
Sample collection. The normal rat was euthanized at 8 weeks after birth. Three N1S1-inoculated rats were euthanized at three weeks after inoculation. After euthanasia, the entire liver was resected. The kidney of the normal rat was resected and used as a positive control of platelet derived growth factor receptor-β (PDGFR-β). Liver and kidney tissues were fixed with 10% formalin. The normal dog was euthanized at 2 years old. After the euthanasia, the entire liver was resected. The liver tissue was fixed with 10% formalin. The dog with HCC underwent a partial resection at two weeks after the first visit. After the surgery, part of the tumor tissue was fixed with 10% formalin.
All formalin-fixed tumor tissues were washed with running tap water to eliminate formalin in the tissues. Formalin-eliminated tissues were then dehydrated through a series of ascending grades of alcohol. Paraffin was then embedded in tumor tissues.
Method of immunohistochemistry. IHC was performed for all tissues of canine and rat groups. Before the IHC, all liver tissues were stained with hematoxylin and eosin (H&E). In H&E staining of the liver tissues, features of malignancy, mitotic figures, anisocytosis, and anisokaryosis, were found (Figure 1). IHC evaluation of the liver tissues was performed in areas where the tumor cells were clearly visible in H&E staining. In the kidney tissue of the normal rat, IHC evaluation was performed around the glomerulus. For IHC, the avidin-biotin-peroxidase complex (ABC) method was used. Formalin-fixed-paraffin-embedded (FFPE) tissues were cut to 4μm in thickness. FFPE tissue sections were mounted on triethoxysilyl propylamine coating slides, deparaffinized, and rehydrated through a series of descending grades of alcohol. Antigens of tissues were recovered by using sodium citrate boiled to 90~100°C. Endogenous peroxidase was deactivated using 3% hydrogen peroxide for 15 min. Non-specific reactions were blocked with diluted horse serum for 1 h at room temperature.
Primary antibodies, monoclonal mouse anti-vascular endothelial growth factor (VEGF) antibody (M7273; DAKO®, Glostrup, Denmark), monoclonal mouse anti-epidermal growth factor receptor (EGFR) antibody (MA5-13269; Invitrogen, MA, USA), monoclonal mouse anti-platelet derived growth factor receptor-α (PDGFR-α) antibody (SC-398206; Santa Cruz Biotechnology, Dallas, TX, USA), monoclonal mouse anti-PDGFR-β antibody (SC-374573; Santa Cruz Biotechnology), and polyclonal rabbit anti-C-kit antibody (A4502; DAKO®), were used. Biotinylated goat anti-rabbit IgG antibody (BA-1000; Vector Laboratories, Burlingame, CA, USA) and biotinylated horse anti-mouse IgG antibody (BA-2001; Vector Laboratories) were used as secondary antibodies. All primary antibodies were diluted at 1:200. Slides were incubated with primary antibodies diluted at 1:200 for 24 h at 4°C and incubated with secondary antibodies diluted at 1:200 for 1 h at room temperature. After incubations, slides were stained with 3,3′-diaminobenzidine and counterstained with hematoxylin. These stained slides were then dehydrated and mounted. As a negative control, the same procedures of IHC except for the incubation of primary antibodies were performed. Using the normal rat kidney, the same IHC procedures were performed for a positive control of PDGFR-β.
Slides were examined with an optical microscope equipped with a digital scanner (Axio Scan.Z1, Zeiss, Germany). Images were captured with a software of the same company (ZEN 3.4 blue edition, Zeiss).
Immunohistochemistry scoring. Two investigators performed blind scoring. Scores of each investigator were averaged. Five spots of each slide at 100X were evaluated. For scoring, a percentage of stained tumor cells (or normal kidney cells) and the intensity of the stain were evaluated for each spot of the slide. Criteria were based on prior studies that used IHC methods for TKIs. Criteria of the percentage of staining consisted of: 0, no staining; 1, up to 25% positive staining; 2, 26 to 50% positive staining; 3, 51% to 75% positive staining; and, 4, >75% positive staining (20). Criteria of the intensity were: 0, no staining; 1, pale staining; and 2, dark staining (21).
For the normal rat, normal dog, and HCC dog groups, the sum of the two scores according to the criteria was used for this study. For the N1S1 rat group, averages of sums of two scores (percentage and intensity) were used for this study.
Statistical analysis. All statistical analyses were performed using GraphPad PRISM (version 9, GraphPad Software, La Jolla, CA, USA). Unpaired t-test and one-way analysis of variance (ANOVA) with subsequent Tukey’s multiple comparison test were used for data analyses. Unpaired t-test was used to compare difference of each factor between normal (the normal rat and normal dog groups) and tumor groups (the N1S1 rat and HCC dog groups). One-way ANOVA with subsequent Tukey’s multiple comparison test was conducted to investigate differences in scores of each factor between all groups. p-Value less than 0.05 was considered statistically significant.
Results
Scores of rat groups. Final scores of the groups are summarized in Figure 1. Images are shown in Figure 2. In the normal rat group, c-kit showed the highest score, followed by VEGF and PDGFR-α. In the N1S1 rat group, VEGF had the highest score, followed by c-kit and PDGFR-α, whereas PDGFR-β and EGFR were not stained in either group. PDGFR-β was stained in the positive control (Figure 3).
Scores of VEGF were almost equal between the normal rat group and that in the N1S1 rat group, showing no significant difference (Figure 3A). Scores of PDGFR-α were significantly lower in the N1S1 rat group than those in the normal rat group (p-value=0.0042, Figure 3B). Scores of c-kit were not significantly different between the normal rat and N1S1 rat groups (Figure 3C).
Scores of dog groups. Final scores of dog groups are summarized in Figure 4. Images are shown in Figure 5. In the normal dog group, VEGF and c-kit showed the highest scores, followed by PDGFR-α. PDGFR-β had the lowest scores. In the HCC dog group, VEGF and c-kit showed the highest scores, followed by PDGFR-β. PDGFR-α had the lowest scores. EGFR was not stained in either group.
Scores of VEGF were not significantly different between the normal dog group and the HCC dog group (Figure 6A). The score of PDGFR-α was significantly lower in the HCC dog group than in the normal dog group (p-value=0.0008, Figure 6B). The score of PDGFR-β was significantly higher in the HCC dog group than that in the normal dog group (p-value=0.0023, Figure 6C). Scores of c-kit were almost equal between the normal dog and HCC dog groups, showing no significant difference (Figure 6D).
Comparison between rat groups and dog groups. Rat groups were compared with dog groups to determine differences in staining patterns between normal groups (the normal rat and normal dog groups) and tumor groups (the N1S1 rat and HCC dog groups). The graphs regarding the scores of all four groups are shown in Figure 7 and Figure 8.
In VEGF and c-kit, there were no significant differences between the scores of the normal and tumor groups in dogs. Also in rats, the scores of the two factors, VEGF and c-kit, were not different between in the normal and tumor groups. The scores of PDGFR-α were significantly lower in the tumor group than those in the normal group of dogs (p-value=0.0001). In the rats, the scores of PDGFR-α in the tumor group were significantly lower than those in the normal group as in the dogs (p-value=0.0438). The scores of PDGFR-β in the dogs were significantly higher in the tumor group than those in the normal group (p-value=0.0002). PDGFR-β was not stained in all rat groups and EGFR was not stained in all four groups.
Based on the results, PDGFR-α and PDGFR-β showed significant differences between the normal dog group and the HCC dog group. The results of PDGFR-α in the rats was consistent with that in the dogs. However, PDGFR-β in the rats was not detected.
Discussion
HCC is the most common primary liver tumor in dogs (1, 2) and the sixth most common cancer in humans (9). In human medicine, orthotopically implanted rat HCC models (N1S1, McA-RH7777) are used for the study of HCC (18, 19). In this study, the N1S1 rat models were compared with the dog with HCC through IHC of the ligand and receptors associated with TKs. Based on the results, VEGF, EGFR, and c-kit, were not significantly different between the normal group and HCC group in dogs. In PDGFR-α and PDGFR-β, there were significant differences between the normal group and HCC group in dogs. PDGFR-α in the rats was consistent with that in the dogs, but PDGFR-β was not stained in the rats of this study.
In this study, VEGF was highly expressed in all four groups, the normal rat, N1S1 rat, normal dog, and HCC dog groups. In other reports, VEGF was expressed in the normal lung, liver, and kidney and had the highest expression in the lung in rats (22). VEGF is also expressed in normal tissues, such as the lung, heart, liver, and renal cortex in dogs (23). Expression rates in the lung, heart, and renal cortex are higher than those in the liver because of vascular permeability (23). The VEGF mRNA levels of HCC tissues were higher than those in the normal tissues of rats (24). In a report comparing canine and human HCC, VEGF was expressed in five out of eight dogs (25). There was no difference between the normal group and tumor group in the dogs of our study. The result in the rats of our study was consistent with that in the dogs. However, the expression of VEGF was not suitable for comparison in this study.
EGFR was not stained in all four groups in this study. In a report about human HCC, EGFR was over-expressed in HCC compared with normal tissues and expression was declined in advanced stages (26). Regarding dogs, there are two controversial reports. One is that EGFR expression in dogs with HCC is lower than that in control groups (27) and the other is that EGFR is expressed in five out of eight dogs (25). In rats, EGFR expression in HCC was also higher than that in regenerative nodules (28). Therefore, additional studies regarding the expression of EGFR in both dogs and rats are required.
PDGFR-α expression in the tumor groups (the N1S1 rat and HCC dog groups) was lower than that in the normal groups in this study. PDGFR-β in the HCC dog group was higher than that in the normal dog group but was not stained in the rat groups. PDGFR-α and PDGFR-β were over-expressed in human HCC (29) and also in rats with HCC induced by methyl-deficient diets (30). The expression of PDGFR-α and PDGFR-β is rarely reported in dogs. In a report, PDGFR-α was positive in three out of four dogs with HCC and PDGFR-β was positive in all four dogs with HCC (31). The intensity and distribution of PDGFR-α are variable but PDGFR-β is stained strongly and widely (31). The result of PDGFR-α in the rats was consistent with that in the dogs in this study. However, overall, the results of PDGFR-α in our study were not consistent with those of prior studies. PDGFR-β expression in the dogs of this study was consistent with that seen in other studies; however, there was no staining in the rats.
PDGFR-β has been found to be expressed in a normal rat kidney (32) by IHC using the same antibodies and procedures. In contrast, PDGFR-β was not stained in the normal rat liver and N1S1 rat liver. Therefore, further studies regarding the differential expression of PDGFR-α between normal and HCC dogs and the expression of PDGFR-β in the rats are needed.
In this study, c-kit was found to be highly expressed like VEGF in all four groups. C-kit expression is positive in human HCC, but the percentage is low in 2.3% to 25.6% of HCC patients (33, 34). C-kit is also over-expressed in one out of four dogs with HCC (31). Furthermore, there is a report about rats in which c-kit positive cells were identified in the HCC-inducing group (35). In this study, the expression of c-kit in the rats was consistent with that in dogs and stained normal groups. Therefore, c-kit expression is not useful for the investigation of dogs with HCC.
Overall, PDGFR-α was significantly different between the normal dog and HCC dog and the results were consistent with the results in rats. However, the results of PDGFR-α in the dogs were contrary to those of other reports. The results of PDGFR-β were consistent with the results of prior studies in the dog groups, but not detected in the rat groups. Therefore, although additional studies are required, PDGFR-α was similar to the N1S1 rat models for the study of canine HCC.
TKRs are proteins that cause tyrosine phosphorylation of other proteins (12). In cancers of human and veterinary medicine, activations of TKRs are caused by mechanisms, mutations, genomic amplification, generation of fusion proteins from chromosomal rearrangements, and autocrine activations (36). Through the abnormal activations of TKRs, uncontrolled cell growth and survival may be induced, and tumor progression can occur (12, 37). In this study, PDGFR-α and PDGFR-β were found to be the most significant factors associated with TKs.
PDGFR is classified into PDGFR-α, PDGFR-β, and PDGFR-αβ (30). Two PDGFRs (PDGFRA and PDGFRB) form dimers (38). PDGF is involved in angiogenesis (39). Therefore, pericyte recruitment to vessels, stimulation of proangiogenic factors, endothelial cell proliferation and migration can occur through PDGF signaling (39). There are some reports about the expression of PDGFR in other canine cancers, such as nasal carcinoma, liposarcoma, and oral melanoma (40-42).
Some studies have used the N1S1 rat model to investigate novel therapies for human HCC including bumetanide, a glycolytic metabolism pathway inhibitor, and boric acid-mediated boron neutron capture therapy (43, 44). Another study studied the combination of sorafenib and doxorubicin-loaded microbubble-albumin nanoparticle complex using this model (45).
However, there is no study using the N1S1 rat models in veterinary medicine. In the future, it is recommended to conduct studies using the tumor rat model to examine the efficacy of targeted anticancer drugs not only on HCC but also in other tumors.
In this study, two points are considered as limitations. First, the liver tissues of the HCC dog used in this study were vacuolated and degenerated. Second, the sample size of each group was small.
In conclusion, the expression of PDGFR-α in the N1S1 rat models in this study was similar to that in dogs with HCC. Further studies on N1S1 rat models for canine malignant tumors are needed.
Acknowledgements
The Authors are grateful to all dogs and dog owners for participating in our investigations. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education (NRF-2020R1A2C2005364).
Footnotes
Authors’ Contributions
SK, YK, S-EK and H-JK designed the experiments. SK and YK carried out the experiments. SK analyzed the data. SK and H-JK wrote the first manuscript draft. H-JK and S-EK supervised the whole study, reviewed, and edited the manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received October 27, 2022.
- Revision received November 14, 2022.
- Accepted November 23, 2022.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).