Effects of selective cyclooxygenase‐2 inhibitor robenacoxib on primary cells derived from feline injection‐site sarcoma

Abstract Feline injection‐site sarcomas (FISSs) are highly invasive malignant mesenchymal neoplasms that arise from injection sites in cats. Although the tumorigenesis of FISSs is still uncertain, there is a consensus that FISS is associated with chronic inflammation caused by irritation of injection‐related trauma and foreign chemical substances. Chronic inflammation can provide a proper microenvironment for tumour development, which has been known as one of the risk factors of tumorigenesis in many tumours. To investigate the tumorigenesis of FISS and screen for its potential therapeutic targets, cyclooxygenase‐2 (COX‐2), an inflammation‐enhancing enzyme, was selected as a target for this study. In vitro experiments using FISS‐ and normal tissue‐derived primary cells and robenacoxib, a highly selective COX‐2 inhibitor, were performed. The results demonstrated that expression of COX‐2 could be detected in formalin‐fixed and paraffin‐embedded FISS tissues and FISS‐derived primary cells. Cell viability, migration and colony formation of FISS‐derived primary cells were inhibited, and cell apoptosis was enhanced by robenacoxib in a dose‐dependent manner. However, susceptibility to robenacoxib varied in different lines of FISS primary cells and was not completely correlated with COX‐2 expression. Our results suggest that COX‐2 inhibitors could be potential adjuvant therapeutics against FISSs.

owing to the high local invasiveness of FISSs, single excisional surgery or radiotherapy is commonly not curative. 14,15 Thus, radiotherapy is the current mainstream adjuvant therapy combined with surgery. 15 Other types of treatments, such as chemotherapy and immunotherapy are rarely administered clinically because of their uncertain therapeutic efficiency.
The pathogenesis of FISSs remains unclear. Currently, the most widely accepted consensus is that FISSs might arise from chronic inflammation of injected sites due to evidence of the presence of aluminium, peripheral multinucleated giant cells, 4,16 peripheral aggregates of inflammatory cells, and a significant epidemiological association between vaccination and sarcomas. Malignant transformation might be induced by excessive inflammatory and immunological responses, which could be stimulated by the adjuvant or other components of the vaccine, followed by abnormal proliferation of fibroblasts and myofibroblasts. 4,17 Chronic inflammation should be regarded as a tumour promoter and carcinogen, 18 such as in human colorectal cancer with intestinal bowel disease, 19 human hepatocellular carcinomas with hepatitis B or C viral infection, 20 and ocular sarcoma with trauma and chronic uveitis in cats. 17 Chronic inflammation involves various immunocytes and factors contributing to the chemotaxis of immunocytes, remodelling of extracellular matrix, removal of pathogens and healing of damaged tissue. 21 Sustained and uncontrolled inflammation leads to a consistent cycle of tissue damage, proliferation and healing 22 ; triggers mutations resulting in neoplastic transformation, including silencing of tumour suppressor genes and activation of oncogenes 23 ; and develops an adequate microenvironment beneficial to oncogenesis, tumour proliferation, invasion and metastasis. 20,22 Cyclooxygenase-2 (COX-2) is an isoform of cyclooxygenase that converts arachidonic acid into prostaglandins, which are known as inflammatory promoters and are considered to collaboratively contribute not only to chronic inflammation but also to an adequate microenvironment for tumorigenesis. 24,25 Additionally, overexpression of COX-2 has been reported in various human tumours, 26 canine [27][28][29] and feline, [29][30][31][32][33] including FISSs. [34][35][36] In human medicine, it has been reported that COX-2 inhibitors could sensitize tumour cells to enhance their response to radiotherapy or chemotherapy. It has also been suggested that chemotherapy drugs or radiotherapy adjuvants with COX-2 inhibitors may provide favourable synergistic responses to antitumor activity. 25 Additionally, previous work showed that nuclear factor kappa B (NF-κB), which is upstream of COX-2, may play a role in FISS tumorigenesis. 37 Accordingly, COX-2 may be a potential therapeutic target for oncogenesis in veterinary medicine. Robenacoxib is a novel COX-2 selective inhibitor with sufficient safety in cats and more intense COX-2 selectivity than meloxicam. 38,39 However, only a few studies have used COX inhibitors for feline tumour treatments, 40,41 and the effect of robenacoxib or COX inhibitors on FISSs has not yet been investigated. This study aimed to investigate the role of COX-2 in the oncogenesis of FISS and to evaluate the therapeutic potential of the selective COX-2 inhibitor, robenacoxib, in FISS-derived primary cells with respect to proliferation, migration, apoptosis and clonogenic ability.

| Cells
Feline injection-site sarcoma primary cell lines were established as previously described, 37 and the FISS cell lines used in the present study were FISS-7 and FISS-12. To obtain normal feline mesenchymal (FM) cells, a normal muscle tissue was collected from a cat euthanized due to other diseases. The culture and passage of FM cells were similar to those of previously described FISS cell lines.

| Immunohistochemistry staining (IHC)
For immunohistochemical staining, 10 micrometre-thick slices were prepared from formalin-fixed and paraffin-embedded (FFPE) tissue blocks of FISSs and normal canine kidneys. 35 The slides were first deparaffinized, rehydrated and then immersed in Trilogy (Cell Marque).
Antigen retrieval was conducted by two rounds of heating at 95°C for 5 min using EZ-Retriver System (BioGenex Laboratories Inc.).
After washing with Tris-buffered saline supplemented with 0.1% Tween® 20 (TBST, Genestar Biotech Inc.), the tissue slides were blocked with 10% of normal goat serum (Dako) diluted in phosphatebuffered saline (PBS) at room temperature (RT) for 30 min and followed with another TBST washing. The COX-2 polyclonal antibody (Cayman Chemical) 35  Finally, after being air-dried and sealed with gum arabic and coverslips, the slides were evaluated using an optical microscope by two pathologists. To quantify the positive cells in the tumour, ten highpower fields were randomly selected and captured, and the pictures were analysed by using ImageJ software (NIH). The ratio was calculated by dividing the positive cell count by the total cell count.

| Immunocytochemistry staining (ICC)
To identify the origin of primary FM cells obtained from muscular tissue and FISS-derived cells, ICCs were incubated with vimentin, desmin and anti-alpha-smooth muscle actin (α-SMA) antibodies. The cells were seeded onto a 96-well plate (4.5 × 10 3 cells/well) and incubated at 37°C for 24 h to reach 80%-90% confluence. After removal of the culture medium, the cells in each well were fixed with 80% acetone at −20°C for 10 min, air-dried at RT and stored at −20°C for subsequent use. One Nikon) by two pathologists. To quantify the positive cells in the tumour, five high-power fields were randomly selected and captured and these pictures were analysed using ImageJ software (NIH). The ratio was calculated by dividing the positive cell count by the total cell count. To detect the expression of COX-2 in FISS and FM cells, ICC was conducted as previously described. The primary antibody was replaced with a COX-2 polyclonal antibody (Cayman) diluted at 1:200 in PBS.

| AlamarBlue cell viability assay for cell viability analysis
The FISS or FM primary cells seeded onto 96-well plates (NUNC, Thermo Fisher Scientific) at a number of 4.5 × 10 3 cells per well were incubated at 37°C for 24 h and treated with culture medium supplemented with robenacoxib at concentrations of 125, 250, 500 and 1000 μM, respectively.
The cells in the negative control group were incubated in a culture medium supplemented with 0.5% DMSO. After the incubation for 24 and 72 h, the AlamarBlue Cell Viability Assay Reagent (G-Biosciences) was used to assess the inhibitory effects of robenacoxib according to the manufacturer's protocol. The 96-well plate was read using the Cytation 7 Cell Imaging Multi-Mode Reader (BioTek Instruments) at 570 and 600 nm, respectively. The percentage reduction was calculated using the protocol provided. The half-maximal inhibitory concentration (IC 50 ) of robenacoxib was calculated by normalized inhibitory response nonlinear regression analysis using PRISM 9 (GraphPad Software).

| Wound healing assay to evaluate the ability of cell migration
Approximately 2 × 10 5 FISS or FM cells per well were seeded in a 24 well-plate (NUNC, Thermo Fisher Scientific) and incubated at 37°C for 24 h to reach 90%-100% confluence. Wounds were generated by scratching the cell monolayers with a pipette tip. Subsequently, the medium was replaced with culture medium supplemented with 125, 250, 500 or 1000 μM of robenacoxib or 0.5% DMSO, respectively.
Each treatment was conducted in triplicates. Photographs were taken at 0 and 24 h, and the width of the wounds was measured using the ImageJ software (NIH). Because there were still several remaining cells after the scratch at 0 h, the edge of cell migration was defined as where the cells reached at least 50% confluence. Ten measurements at different sites were performed to obtain the average widths of the wounds at each treatment and time point. The data are presented as a percentage compared to the average value at 0 h for each treatment. Cells treated with culture medium added with 0.5% DMSO were designated as the vehicle control. All treatments were conducted in triplicates. After a three-day incubation period, the medium was removed and replaced with a drug-free culture medium every 3-5 days until cell colonies could be detected by the naked eye (more than The results were evaluated by plating efficiency (PE) and survival fraction (SF). 42 Plating efficiency was defined as the percentage of colony formation and number of cells seeded in the well, and SF was defined as the number of colonies formed after treatment divided by the number of cells seeded in multiple PE.

| Terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) assay for cell apoptosis
To evaluate the effect of robenacoxib on apoptosis, the DeadEnd™ Fluorometric TUNEL system (Promega, Madison) was used.

| Statistical analysis
Statistical analysis was performed using SAS version 9.4 (SAS Institute, Cary, North Carolina, United States). Group comparisons between the controls and treatments were conducted using a oneway analysis of variance with the Scheffé test. Statistical significance was defined as a p-value <0.05. The IC 50 and its inhibitory curves were analysed using Prism 9 (GraphPad Software).

| Immunophenotypes in primary cells derived from FISS and normal feline tissues via ICC staining
Three isolates of primary cells were used in the present study, including cells derived from normal feline tissues and FISSs.
Histopathologically, the original masses of both FISS cells were categorized as grade III soft tissue sarcomas according to the grading system published by Dennis et al. 43 ; the detailed information is listed in Table S1. Immunocytochemical staining for vimentin, desmin and α-SMA was performed and the results are shown in Figure 1.

| COX-2 expression in FISS-7 and FISS-12
To evaluate the expression of COX-2 in FISS cells, ICC and IHC staining and western blotting against COX-2 were performed ( Figure 1). In   Figure 3B. This shows that FISS-12 cells had the highest COX-2 expression level among the three primary cell lines.

| Safety of COX-2 inhibitor robenacoxib in FISS and FM cells
To investigate the role of COX-2 in the pathogenesis of FISSs, the effect of robenacoxib on FISS cells was studied to investigate the present study. Drug safety was evaluated by performing the AlamarBlue cell viability assay and presented as IC 50 . The IC 50 values of each primary cell line after 24 and 72 h of treatment are shown in Figure S1 and Table 1  However, there was no obvious difference in the IC 50 and cell viability between FISS-12 and FM cells.

| Cell migration effects of robenacoxib on FISS and FM cells, as measured by a wound healing assay
A wound healing assay was performed to evaluate the effect of robenacoxib on the migration of different primary cells, and the results are shown in Figure 4 and Table S5-

| Effect of robenacoxib on the cell clonogenesis in FISS and FM cells
To study the effect of robenacoxib on cell colony formation, clonogenic assays of FISS-7, FISS-12 and FM cells were performed, and the results are displayed in Figure 5 and Table S8. The results showed that all lines of primary cells shared a similar pattern with higher concentrations of robenacoxib, including 500 and 1000 μM, exhibiting a significant inhibitory effect on colony formation (p < 0.05). As for cells treated with 0, 125, or 250 μM robenacoxib, there was no significant difference in each cell when compared to that of their own medium control (p > 0.05). At 500 μM, FISS-7 and FM cells tended to have more colonies than FISS-12 cells, suggesting that the clonogenesis of FISS-12 may be more sensitive to robenacoxib than FISS-7 and FM cells.

| Apoptosis rate of FISS and FM cells detected using TUNEL assay
To evaluate the effects of robenacoxib on apoptosis in FISS-7, FISS-12 and FM cells, TUNEL assay was performed and the results are shown in Figure 6 and Table S9. Generally, cell apoptosis was more prominent in treatments with higher concentrations of robenacoxib, especially 500 and 1000 μM, in a dose-dependent manner, but without statistically significant differences (p > 0.05). However, comparing the results of different cells treated with similar drug concentration, the apoptotic rates of FISS-12 cells at all concentrations were higher than those of FISS-7 and FM cells. well. 28 In canines, COX-2 protein overexpression has been identified in mesenchymal and epithelial tumours. [44][45][46][47][48][49][50][51] Additionally, in previous studies, the expression of COX-2 detected by immunohistochemistry varied among FISSs, but the positive rates of COX-2 expression in the FISS-affected cats ranged between 0%-97%. 30,[34][35][36] It has also been reported that COX-2 expression could be detected in neoplastic cell lines of canine transitional cell carcinoma and feline oral squamous cell carcinoma by immunohistochemical staining, suggesting that COX-2 inhibitors could be potential candidates for anticancer agents directly affecting neoplastic cells in dogs and cats. 52 In our study, the immunoreactivity of COX-2 in FFPE tissues

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data in this study are available from the corresponding author upon reasonable request.