The human colon cancer HCT116 cell line (cat. no. 10247) and CRC HT-29 cell line (cat. no. 30038) originating from colorectal adenocarcinoma (60) were purchased from the Korean Cell Line Bank (Korean Cell Line Research Foundation) and were authenticated by Procell Life Science & Technology Co., Ltd. using STR analysis. The two cell lines were cultured in RPMI-1640 (cat. no. 23400-021; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (cat. no. S001-07; Welgene Co., Ltd.), 1% penicillin-streptomycin (cat. no. P4333-100ml; Sigma-Aldrich; Merck KGaA) in a humidified chamber at 37°C under 5% CO₂. UA (purity, >95%; cat. no. 10072) was purchased from Cayman Chemical Company. DOX hydrochloride (purity, 98–100%; cat. no. D1515-10MG) and SC79 (AKT activator) (purity, ≥97%; cat. no. 305834-79-1) were purchased from Sigma-Aldrich; Merck KGaA. LY294002 hydrochloride (AKT inhibitor) (purity, ≥98%; cat. no. 934389-88-5) was purchased from LKT Laboratories, Inc. Following pre-treatment with AKT inhibitor LY294002 (1 µM) for 2 h or AKT activator SC79 (5 µM) for 2 h, UA + DOX combination treatment was applied for 48 h.

A total of 1×10⁴ HCT116 cells or HT-29 cells were seeded onto 96-well plates and treated with DOX (500 nM, 1, 1.5, 2 µM), UA (5, 10, 15, 20 µM) or DOX (1.5 µM) with or without UA (15 µM) for 48 h at 37°C under 5% CO₂. In the control group, no pretreatment was added. To the DMSO group, similar amounts of DMSO were added as to the DOX group (0.2 µl per well). In the combination group, DOX (1.5 µM) and UA (15 µM) were administered simultaneously. Subsequently, 10 µl EZ-Cytox (DoGenBio Co., Ltd.) was added and incubated for 2 h at 37°C. Finally, the absorbance at 450 nm was assessed using a microplate reader (Epoch Microplate Spectrophotometer; BioTek Instruments, Inc.; cat. no. VT 05404). Cell proliferation is presented as the percentage relative to untreated cells (control group).

Briefly, a base layer containing 1% agar (micro agar, cat. no. M1002.1; Duchefa Biochemie B.V.) and a top layer containing 0.7% agar (Agarose SPI, cat. no. A1203.01; Duchefa Biochemie B.V.) were inoculated with 1×10⁵ cells per well. The medium, supplemented with UA (15 µM) and/or DOX (1.5 µM), was changed three times/week and 6-well plates were incubated for 14 days at 37°C under 5% CO₂. Colonies were defined as >30 cells, were counted manually and were observed using an IX71 fluorescence microscope (Olympus).

HCT116 and HT-29 cells (2.5×10⁵) were seeded onto 6-well plates and treated for 48 h at 37°C with UA (15 µM) and/or DOX (1.5 µM). Propidium iodide (cat. no. P4170; Sigma-Aldrich; Merck KGaA) and Annexin-V-FITC Assay kit (BD Biosciences) were used to stain the cells for 30 min at room temperature in the dark according to the manufacturer's protocols. BD Accuri C6 Flow Cytometer (BD Biosciences) was used to assess the cell apoptosis. Acquisition and analysis of the data were performed using BD Accuri CFlow software (version, 1.023.1; BD Biosciences). Cells that were considered viable were FITC Annexin V and PI negative; cells that were in early apoptosis were FITC Annexin V positive and PI negative; and cells that were in late apoptosis or already dead were both FITC Annexin V and PI positive.

HCT116 and HT-29 cells (2×10⁶) were seeded onto a 6-well plate with RPMI-1640 containing 10% FBS. In order to reduce the risk of cell proliferation, 90–100% confluent HCT116 or HT-29 monolayers were incubated with RPMI-1640 containing 1% FBS, the DNA synthesis inhibitor mitomycin C (5 µg/ml) (cat. no. 10107409001; Sigma-Aldrich; Merck KGaA) 6 h prior to the scratch assay (61–63). A 200-µl pipette tip was used to scrape a straight line and cells were treated with UA (15 µM) and/or DOX (1.5 µM) for 48 h at 37°C. The wound width in five random views was quantified at 0, 24 and 48 h to assess the effect of UA (15 µM) and/or DOX (1.5 µM) on cell migration. Images were captured with an IX71 inverted fluorescence microscope (Olympus Corporation). Migration was calculated as follows: Migration (%)=[(0 h mean scratch distance −24 or 48 h mean scratch distance)/0 h mean scratch distance] ×100. The mean scratch distance can be calculated between the scratches in each of the five random views. The cell migration rate was assessed using Image-Pro Plus 6.0 software (Media Cybernetics, Inc.).

Western blot analysis was performed as described previously (64). Briefly, proteins were extracted from HCT116 cells, HT-29 cells and animal tumor tissues using RIPA lysis and extraction buffer containing protease and phosphatase inhibitors (Intron Biotechnology, Inc.). The supernatant was centrifuged at 16,100 × g for 30 min at 4°C. After centrifugation, protein quantification was performed using Pierce™ Rapid Gold BCA Protein Assay kit (cat. no. A53225; Thermo Fisher Scientific Inc.). The standard curve was used to determine the protein concentration of each sample. An equal amount of total protein (30 µg/lane), and 10 µl DokDo-MARK Broad-Range Protein Ladder (cat. no. EBM-1032; ELPIS-BIOTECH, Inc.) or 10 µl Enhanced 3-color High Range Protein Marker (cat. no. PM2610; SMOBIO Technology, Inc.), were separated by SDS-PAGE on 8 and 10% gels and transferred onto 0.45 µm PVDF membranes (cat. no. 1620174; Bio-Rad Laboratories, Inc.). The membranes were blocked using 5% skimmed milk in TBS −0.1% Tween-20 (cat. no. 9005-64-5; Sigma-Aldrich; Merck KGaA) for 1 h at room temperature. The membranes were then incubated overnight at 4°C with primary antibodies purchased from Cell Signaling Technology, Inc. as follows: GAPDH (1:1,000; cat. no. 2118S), cleaved-poly (ADP-ribose) polymerase 1 (PARP; 1:1,000; cat. no. 9541S), cleaved caspase-9 (1:1,000; cat. no. 9501S), PARP (1:1,000; cat. no. 9542S), caspase-9 (1:1,000; cat. no. 9502S), E-cadherin (1:1,000; cat. no. 5296S), cyclin D1 (1:1,000; cat. no. 2922), cyclin-dependent kinase (CDK) 4 (1,000; cat. no. 2906), CDK6 (1,000; cat. no. 3136s), Akt (1:1,000; cat. no. 9272), phosphorylated (p)-Akt (1:1,000; cat. no. 9271S), Gsk3β (1:1,000; cat. no. 12456S), p-Gsk3β (1:1,000; cat. no. 9322S), Mst1 (1:1,000; cat. no. 3682S), Mst2 (1:1,000; cat. no. 3952S), salvador family WW domain containing protein 1 (Sav1; 1:1,000; cat. no. 3507), MOB kinase activator 1 (Mob1; 1:1,000; cat. no. 3863S), p-Mob1 (1:1,000; cat. no. 8699), Yap (1:1,000; cat. no. 4912) and p-Yap (1:1,000; cat. no. 4911S). Primary antibodies against Ras association domain family member 1 (Rassf1A; 1:1,000; cat. no. sc-58470), urokinase-type plasminogen activator (uPA; 1:1,000; cat. no. sc-14019), matrix metallopeptidase 9 (MMP-9; 1:1,000; cat. no. sc-6840), c-Myc (1:1,000; cat. no. sc-40) and connective tissue growth factor (CTGF; 1:1,000; cat. no. sc-14939) were purchased from Santa Cruz Biotechnology, Inc. Membranes were then incubated with anti-mouse (1:2,000; cat. no. 7076; Cell Signaling Technology, Inc.), anti-goat (1:2,000; cat. no. AF109; Novus Biologicals, Inc.) or anti-rabbit (1:2,000; cat. no. 7074S; Cell Signaling Technology, Inc.) horseradish peroxidase-conjugated secondary antibodies for >2 h at 4°C. The enhanced HRP substrate luminol reagent was used to visualize the bands (Amersham; Cytiva). Immunoreactivity was detected using Fusion Fx7 (Vilbe, Lourmat). Films were exposed at multiple time points to ensure images were not saturated. ImageJ (version 1.52; National Institutes of Health) was used to semi-quantify the blot images.

The animal experiments were approved by the Institutional Animal Care and Use Committee of Jeonbuk National University (approval no. CBNU2017-0001; Jeonju, South Korea) under National Institutes of Health guidelines (65). The present study was performed in compliance with the Animal Research: Reporting of In Vivo Experiments guidelines 2.0 (66). The present study was performed according to the guidelines on the welfare and use of animals in cancer research (67). Male SPF/BALB/c nu/nu immunodeficient mice were purchased from Orient Bio Inc. Athymic nude mice (age, 4 weeks; weight, 20±3 g; n=20) were acclimated to the animal housing conditions for 1 week. Mice were housed in an experimental animal facility under standard laboratory conditions (20–25°C, 40–60% humidity, 12-h light/dark cycle) with free access to food and water. HCT116 cells (1×10⁷) in 100 µl Matrigel (Corning Matrigel Basement Membrane Matrix; cat. no. 356234; Corning, Inc.) were injected into the flank area of the mice. After inoculation, HCT116 tumor-bearing animals were assigned to four groups (n=5/group) as follows: i) vehicle; ii) UA; iii) DOX; and iv) UA + DOX and housed in individually ventilated cages with litter and nesting material. The mice in the vehicle group were administered 20 µl sterile DMSO twice weekly. The mice in the UA group were treated with 10 mg/kg/day UA in 100 µl PBS. The mice in the DOX group were administered 2 mg/kg DOX in 20 µl DMSO twice weekly. The mice in the DOX + UA combination group were administered 2 mg/kg DOX in 20 µl DMSO twice weekly and 10 mg/kg/day UA in 100 µl PBS. All agents were administered via intraperitoneal injection. The average volume of the tumors was calculated as follows: Tumor volume=width² × length/2. The body weight of mice was monitored every 2 days. Mouse welfare was assessed daily and palpable tumors were assessed every other day. All mice could reach food and water, and all had a fairly normal gait. Mice moved freely in the cage with normal body and tail position and in a forward motion; the abdomen of the mice was soft with a mild resistance to pressure when touched and abdominal distention was not observed in the mice following injection of the drugs. All mice demonstrated a slightly curved back and did not have a hunched position when sitting or during rest. One mouse had bite wounds on the tail and back; however, blood stains were absent in the cage and the mouse was separated to an individual cage on day 13. Wounds were locally disinfected daily until healed and the mouse was isolated until the end of the procedure. All mice had no ocular or nasal discharge. When the skin of the neck of the mice was pinched, one mouse in the control group demonstrated poor skin turgor and the tumor diameter of this mouse reached 2 cm on day 22. The poor turgor indicated that this mouse was dehydrated and all mice were sacrificed on day 22. Mice were placed in a chamber with 4% isoflurane (cat. no. 1349003; Sigma-Aldrich; Merck KGaA) until animals lost consciousness and CO₂ (50% of the chamber vol/min) was used to euthanize the mice. Death was confirmed by loss of heartbeat, toe reflex, muscle tone, breathing and corneal reflex, and rigor mortis; subsequently, tumor tissues were harvested.

Serum concentrations of aspartate aminotransferase (AST; cat. no. 10103; Asan Pharmaceuticals Co., Ltd.), alanine aminotransferase (ALT; cat. no. 10102; Asan Pharmaceuticals Co., Ltd.), blood urea nitrogen (BUN; cat. no. K024-H1; Arbor Assays LLC Co., Ltd.) and creatinine (cat. no. KB02-H1; Arbor Assays LLC Co., Ltd.) were determined using commercial kits.

At the end of the animal experiment, mouse tumor tissue samples were embedded in 10% formaldehyde at 4°C for 24 h, sliced into 4-µm sections and stained using hematoxylin and eosin (H&E) following a standard protocol for histopathological examination (68). For immunohistochemistry staining, sections from mouse tumor tissue were prepared according to the manufacturer's protocol using the Rabbit HRP/DAB Detection IHC kit (cat. no. ab64261; Abcam). Samples were deparaffinized with xylene, rehydrated with a descending ethanol series l (100, 95, 90, 80 and 70% ethanol) at room temperature for 5 min and subjected to antigen retrieval by boiling samples in decreasing concentrations of EDTA buffer (pH 9.0) at 120°C for 2 min, followed by cooling to room temperature for 30 min. Slides were blocked using hydrogen peroxide blocking buffer (supplied in the kit) for 15 min to quench endogenous peroxidase and 10% goat serum (cat. no. 16210064; Thermo Fisher Scientific, Inc.) at room temperature for 30 min each. Slides were washed with PBS and incubated with rabbit anti-Ki-67 (1:100; cat. no. PA5-16785; Invitrogen; Thermo Fisher Scientific, Inc.) primary antibodies overnight at 4°C and biotinylated goat anti-rabbit IgG and streptavidin peroxidase (cat. no. ab64261; Abcam) at room temperature for 2 h. Slides were washed with PBS and counterstaining was performed using 100 µl diaminobenzidine at room temperature for 1 min and sufficient hematoxylin to cover the slides at room temperature for 1 min. Histopathological assessment and immunohistochemistry staining analysis were performed using a light microscope.

In vitro experiments were performed in triplicate. The number used for animal experiments was n=5. Data analysis was performed using GraphPad Prism 8 software (GraphPad Software, Inc.). One-way ANOVA followed by Tukey's post hoc test was used to assess differences among ≥3 groups. The data are presented as the mean ± SEM. P<0.05 was considered to indicate a statistically significant difference.

In the present study, the anti-proliferative effects of UA and DOX were assessed by cell proliferation assay to evaluate cytotoxicity. The UA and DOX concentrations used in the in vitro experiments were based on the cell proliferation assay results. Following 48 h of exposure, UA or DOX markedly inhibited the cell proliferation of both HCT116 and HT29 cells in a concentration-dependent manner compared with the control (Fig. S1). DOX (1.5 µM) treatment of HCT116 cells caused significant cell proliferation suppression compared with in the control group (Fig. 1A). The UA (15 µM) + DOX (1.5 µM) combination treatment significantly inhibited cell proliferation compared with in the UA and DOX groups. HT-29 cell proliferation was significantly decreased (75–85% after 48 h) by co-treatment compared with either agent alone. These data suggested that cell proliferation was significantly suppressed by the combination of UA + DOX compared with either agent alone. Subsequently, a soft agar colony formation assay was performed using CRC cells to assess the effect of combination treatment on colony formation. UA + DOX markedly decreased the size and significantly decreased the number of colonies compared with each drug alone in both CRC cell lines (Fig. 1B).

Flow cytometry was used to assess the apoptotic status of cells following treatment with different drugs. It has been reported that the sub-G₁ phase is characterized by apoptosis induction (69). The proportion of cells in sub-G₁ phase was significantly increased following treatment with the combination of 15 µM UA + 1.5 µM DOX compared with the control and either agent alone (Figs. 1C and S2A). PI/FITC assay results demonstrated that the combination treatment significantly induced apoptosis compared with the control and either agent alone (Fig. 1D). Furthermore, western blotting demonstrated markedly decreased caspase-9, significantly decreased PARP, and significantly increased cleaved caspase-9/caspase-9 and cleaved-PARP/PARP protein expression levels following treatment with the combination of 15 µM UA + 1.5 µM DOX compared with the single treatments in HCT116 and HT-29 cells (Fig. 1E). These data were consistent with the flow cytometry results. The present study indicated that combination treatment triggered the apoptosis of CRC cells.

Distant cancer cell metastasis is a key step that includes migration, invasion and EMT, which represents an advanced malignancy stage (70). The migratory ability of CRC cells, as demonstrated using wound healing assay, was significantly decreased following treatment with the combination of UA + DOX compared with UA or DOX alone (Fig. 2A). Investigation of EMT-associated regulators was performed using western blotting. The protein expression levels of E-cadherin, an epithelial marker, were markedly increased, whereas the protein expression levels of mesenchymal markers (MMP-9 and uPA) were significantly decreased in the combination groups in both cell lines compared with the control and either agent alone (Fig. 2B). These data demonstrated that combination treatment with UA + DOX may have inhibited cell migration via disruption of the EMT process.

In the present study, flow cytometry was used to assess cell cycle progression. The results suggested that UA did not affect G₁ phase in HCT116 and HT29 cell lines, whereas cell accumulation in G₁ phase was markedly increased following DOX treatment compared with in the control group. Furthermore, combination treatment with UA + DOX in HCT116 and HT-29 cells further induced G₁ phase arrest, followed by a marked decrease in the number of cells in G₂ phase compared with in the control group (Figs. 3A and S2B). Western blotting demonstrated that cyclin-dependent kinase (CDK)4/6 and cyclin D1 protein expression levels, which are essential for DNA synthesis (71), were significantly downregulated following combination treatment compared with in the control group (Fig. 3B). Collectively, these data indicated that UA further enhanced DOX-mediated cell cycle arrest in the G1 phase in HCT116 and HT-29 cell lines.

Growing evidence has indicated that overactive PI3K/Akt signaling serves a crucial role in cell migration and proliferation, as well as apoptosis inhibition (72–74). In the present study, the protein expression levels of p-Akt, p-Gsk3β and c-Myc were assessed using western blotting to evaluate the effects of UA and DOX treatment on the Akt signaling pathway. The combination treatment of UA + DOX significantly decreased the protein expression levels of p-Akt and c-Myc, which is downstream of the Akt signaling pathway (75), compared with the control, whereas the total level of Akt remained unchanged (Fig. 4). Consistently, p-Gsk3β (S9) expression levels were markedly decreased by combination treatment compared with UA or DOX alone (Fig. 4). Collectively, these results demonstrated that combination treatment with UA + DOX exerted anticancer effects by inhibiting the Akt signaling pathway.

Guo et al (76) reported that Rassf1A directly binds to Mst1 and Mst2, inhibiting tumorigenesis by activating the Hippo signaling pathway. Therefore, the effect of combined treatment on the expression levels of Rassf1A and Hippo signaling pathway-associated proteins were assessed using western blotting. Mst1 and Mst2 activate downstream tumor suppressor Sav1, which also serves a key role in the Hippo pathway (77). The expression levels of Rassf1A, Mst1/2, Sav1 and the ratio of p-Mob1/Mob1 were all significantly elevated following combination treatment compared with UA or DOX treatment alone in both CRC cell lines (Fig. 5A and E). Furthermore, the protein expression levels of p-Yap, which is the inactivated form of Yap, were significantly induced by combined treatment compared with the control. However, Yap protein expression levels were significantly decreased compared with the control. Furthermore, combination treatment significantly suppressed the protein expression levels of CTGF, which is a target of Yap (78), compared with the control (Fig. 5B and F). Collectively, these results indicated that UA + DOX combined treatment may inhibit Yap-induced CRC progression by activating the Hippo signaling pathway. Subsequently, the effects of an Akt inhibitor (LY294002) and activator (SC79) on the Hippo pathway were assessed to evaluate the crosstalk between the Hippo and Akt pathways. LY294002, a PI3K inhibitor, restricts downstream Akt signaling activity (79). LY294002 markedly decreased the ratio of p-Akt/Akt while activating the Hippo pathway compared with control group (Fig. 5C and G). The expression levels of Hippo pathway proteins Rassf1A, Mst1 and Sav1 demonstrated no significant difference following SC79 treatment. Notably, combination treatment of LY294002 + UA + DOX significantly inactivated Akt signaling activity as demonstrated by significantly increased protein expression levels of Rassf1A, Mst1/2 and Sav1 compared with the UA + DOX treatment group. LY294002 + UA + DOX treatment also enhanced the expression of p-Yap while significantly increasing the ratio of p-Yap/Yap and decreasing the expression of its target CTGF (Fig. 5D and H). Collectively, these results demonstrated that combination treatment of LY 294002 + UA + DOX resulted in the inhibition of Akt activity together with Hippo signaling pathway activation.

Based on the aforementioned in vitro data, the anti-CRC effects of UA and DOX were evaluated in xenograft tumor tissue in vivo. A concentration of 15 µM UA was used to assess whether a low concentration of UA enhanced the action of DOX. DOX concentration was selected using half maximal inhibitory concentration values (data not shown). The data from in vitro experiments suggested that cell proliferation was significantly suppressed by the combination of 15 µM UA + 1.5 µM DOX compared with UA or DOX treatment alone. Certain previous studies reported that administration of UA (12.5 mg/kg daily) or UA (10 mg/kg for 5 consecutive days) significantly inhibited tumor growth in CRC xenograft mouse models (80,81). Our previous study also demonstrated that UA treatment (10 mg/kg daily) suppressed tumor growth in an esophageal cancer xenograft mouse model, and did not impair liver or kidney function (64). Other previous studies have reported that DOX (2 mg/kg/week) or DOX (5 mg/kg every 2 days for a total of five times) treatment suppressed mouse CRC xenografts in vivo (82,83). Therefore, based on the concentration used in other studies, experimental dosages of 10 mg/kg UA for 5 consecutive days and 2 mg/kg DOX twice a week were selected for the CRC xenograft mouse model. Previous studies have recommended human DOX dosage to be 60–75 mg/m² as a single intravenous injection administered at 21-day intervals, with a lifetime cumulative dose limit of 550 mg DOX/m² body surface area (84,85). For body surface normalization, human values were corrected for the mean body surface area of 1.62 m² (60 kg) and mouse values were corrected for the mean body surface area of 0.07 m² (20 g) (86). Therefore, the mouse equivalent dose was determined using the following equation: Mouse equivalent dose (mg/kg)=human dose (mg/kg) × (human K_m/mouse K_m) where the K_m factor for each species was constant [human K_m, 37; mouse K_m, 3; K_m is estimated by dividing the average body weight (kg) of species to its body surface area (m²)] (87). Based on recommended human DOX dosage, the mouse equivalent dose was 19.98-24.975 mg/kg. In the present study, low dose DOX (2 mg/kg, twice/week) treatment with a cumulative dose of <10 mg/kg notably inhibited tumor growth in the CRC xenograft mice model (data not shown).

In the present study, levels of the biochemical markers ALT, AST, BUN and creatinine were assessed in mouse serum to evaluate the organ toxicity of UA and DOX. UA and DOX treatment did not significantly affect the ALT, AST, BUN or creatinine level between the four groups (Fig. 6A-D). This result indicated that the doses of UA and DOX did not impair liver or kidney function. As presented in Fig. 6E, the weight of most of the tumor-bearing mice remained unchanged throughout the study period, with only three mice losing <5% of their weight in the vehicle and UA treatment groups. UA and DOX treatment alone significantly decreased tumor weight (Fig. 6F), tumor size (Fig. 6G) and tumor volume (Fig. 6H) compared with in the vehicle group. UA and DOX treatment demonstrated notable inhibition of tumor growth compared with the control group in the xenograft model. Furthermore, a marked different histology was demonstrated using H&E staining (Fig. 6I). Aggressive tumor cell proliferation with a high nucleus/cytoplasm ratio was observed in the control group, whereas the combination of UA + DOX treatment induced lymphocyte infiltration around tumor cells and increased the areas of apoptosis (Fig. 6I). Moreover, immunohistochemistry staining demonstrated that expression of the proliferation marker Ki-67 in the UA + DOX treatment group was lower than that in the control group, which suggested that the combination of UA + DOX suppressed tumor growth and development in vivo. Tumor samples were harvested and western blot analysis was performed to assess the molecular mechanisms underlying the effects of combination treatment in vivo. The combination treatment significantly decreased the p-Akt/Akt ratio compared with UA or DOX alone while markedly activating the Hippo signaling pathway compared with UA or DOX alone (Fig. 6J-L). Collectively, these in vivo results demonstrated that UA augmented the antitumor effects of DOX via inactivation of Akt and activation of the Hippo signaling pathway in the mouse xenograft model.