Human breast cancer cell lines (MCF-7 and MDA-MB-231) and murine breast cancer cell line 4T1 were purchased from the cell bank of the Chinese Academy of Sciences (Shanghai, China). The MCF-7 kinase knockout library cells, used CRISPR-Cas9 technology to target 507 kinases, were purchased from Focus Biology (Nanchang, China). All cells were cultured at 37°C in DMEM supplemented with 10% fetal bovine serum (FBS, Biological Industries, Beit-Haemek, Israel). We have successfully isolated and obtained Orf virus (NA1/11 strain). And the method of virus propagation and purification refers to our team’s previous research (Li et al., 2012).

PAK4 inhibitor PF-3758309 was purchased from Selleck Chemicals (Shanghai, China). In the in vitro experiment, the mother liquor of PAK4 inhibitor was dissolved in DMSO and diluted with the basic medium. In the in vivo experiment, it is dissolved by the ratio of 5% DMSO + 30% PEG300 + 65% normal saline (Murray et al., 2010; Bradshaw-Pierce et al., 2013; Pitts et al., 2013).

Total RNA was extracted through Cell Total RNA Isolation Kit (FOREGENE, Chengdu, China) and reverse transcribed to cDNA using a reverse transcription kit (TOYOBO, Shiga, Japan). Quantitative real-time PCR was carried out with SYBR master mixture (TOYOBO, Shiga, Japan) on ABI 7500 Real-Time PCR platform following the manufacturer’s instructions (Thermo Fisher Scientific, MA, United States). We use a 3-step cycle protocol which was constructed following Toyobo’s guidelines: 1. Pre-denaturation in 95°C for 1 min; 2. 40 cycles were performed, denaturation in 95°C for 15 s, annealing in 65°C for 15 s and extension in 72°C for 45 s. The primers used were shown in Table 1. The results were presented as the mean ± SEM from three experiments.

Cells were prepared in 96-well plates at the concentration of 3 × 10³ cells/well. Cells were infected with ORFV (MOI = 0, 1, 2, 5) or treated with PF-3758309 (20 nM) for various time points. Cell viability was detected by CCK-8 assay using CCK-8 solution (Sangon Biotech, Shanghai, China) in accordance with the manufacturer’s instruction.

Colony forming ability of MDA-MB-231, MCF-7 and 4T1 cells infected with ORFV (MOI = 5) or PBS was evaluated by colony formation assay. The cells grown in log phase were seeded on a 6-well plate at a density of 3,000 per well. After ORFV treatment, cells were fixed with absolute ethanol and stained with 0.1% crystal violet solution. The experiments were repeated at least three times.

The flow cytometry assay was performed to detect cell cycle and apoptosis of tumor cells. The breast cancer cells infected by ORFV (MOI = 5) or PBS for 24 h of detecting cell cycle, and undergone various treatments for 48 h of detecting apoptosis. The collected cells were processed using Cell Cycle Staining Kit and Annexin V-FITC/PI Apoptosis Detection Kit (Lianke Biotech, Hangzhou, China) according to the manufacturer’s instructions and detected using Guava Flow Cytometer (Millipore, MO, United States).

Cells were lysed in 1 × SDS protein loading buffer to obtain a total protein sample. The proteins separated on a 12% SDS polyacrylamide gel and blotted on a PVDF membrane. After blocking the 5% skim milk powder at room temperature for 2 h, the primary antibody was incubated overnight at 4°C, the secondary antibody was incubated at room temperature for 1 h, and then the PVDF membrane was washed 5 times with TBST. And visualized with ECL chemiluminescence reagent (Yeasen, Shanghai, China) using GeneGnome Chemiluminescence Imaging System (Gene, Hong Kong, China).

Female BALB/c mice (Laboratory Animal Center of the Southern Medical University, Guangdong, China) were used in experiments at 6 weeks of age. All animal testing procedures were strictly in accordance with the regulations of the Laboratory Animal Care. In the allograft tumor model, 4T1 cells (10⁶ in 100μL PBS) were subcutaneously injected into BALB/c mice. When tumors average volume about 50 mm³ (tumor volume (mm³) = 0.5 × length × width²), mice were randomly divided into four groups: ORFV group was received ORFV (10⁷ pfu) via intratumoral injection and normal saline by oral gavage; PAK4i group was received PBS via intratumoral injection and PF-3758309 (20 nM) by oral gavage; Combo group was received ORFV (10⁷ pfu) via intratumoral injection and PF-3758309 (20 nM) by oral gavage; and Mock group was received PBS via intratumoral injection and normal saline by oral gavage. At the end of the experiment, mice were sacrificed by cervical dislocation. The organs and tumor tissues were isolated and fixed in 10% buffered formalin for subsequent experiments.

Tumor tissues and organs isolated from the mice were fixed and embedded in paraffin. Paraffin sections were made by Servicebio company (Wuhan, China). The slides of the tumor tissues were used for IHC staining of cleaved caspase 3 (Cell Signaling Technology, MA, United States), and IF staining of Ki67 (Proteintech, IL, United States).

Three days after ORFV intratumoral injection treatment, the mice were sacrificed and the spleen and tumor tissues were isolated. HE staining and CD3 IHC staining of spleen were completed by Servicebio company (Wuhan, China). Total RNAs were extracted from the tumor tissues and subjected to qPCR to determine the expression of CCL3, CCL5, CCL19, and CXCL12.

After 72 h of infection of MCF-7 kinase knockout library cells with ORFV (MOI = 5), the surviving cells were collected and re-cultured. Genomic DNA was extracted until the number of cells was sufficient. Genomic DNA of untreated MCF-7 kinase knockout library cells were extracted as a control.

Using the extracted genomic DNA as a template, PCR is performed with CRISPR/Cas 9 specific primers (Table 1) to obtain amplified products containing sgRNA specific sequences. The PCR amplified product was recovered and purified and sent to IGE Biotechnology (Guangzhou, China). for next-generation sequencing.

Design siRNAs corresponding to the candidate kinase, and use siRNA-NC as a control (Table 2). After 24 h of transfection with siRNA, MCF-7 cells were infected with ORFVGFP, genetically engineered ORFV carrying green fluorescence, for the next 24 h. Collect cells to detect green fluorescence by flow cytometry to estimate the infection efficiency of ORFV.

A Matrigel precoated chamber (Corning, MA, United States) was used to analyze cell invasion ability. Cells were exposed to ORFV or PF-3758309 or combined treatment for 24 h. The cells in serum-free medium are loaded into the upper chamber, and the lower chamber is loaded with medium containing FBS. After 24 h of incubation, non-invasive cells were removed, and invasive cells were stained with crystal violet. Finally, count the number of invaded cells through microscope.

Data are presented as mean ± SEM. Statistical analyses were performed with GraphPad PRISM (6.0) software. Significant differences between two groups were compared using a Student’s t-test (two-tailed). A one-way ANOVA was used in multiple comparisons. The Chou-Talalay method for drug combination is based on the median-effect equation, which provides the theoretical basis for the combination index (CI)-isobologram equation that allows quantitative determination of drug interactions, where CI < 1, = 1, and > 1 indicates synergism, additive effect and antagonism, respectively (Zhang et al., 2015). P-values of < 0.05 were considered significant, and *p < 0.05, **p < 0.01, and ***p < 0.001.

We first infected three different breast cancer cells, including MDA-MB-231, MCF-7, and 4T1, with ORFV at MOI of 1. For the purpose of detecting ORFV replication, we compared the mRNA expression levels of ORFV core gene ORFV035 at different time points after ORFV infection (Figure 1A). The mRNA expression of ORFV035 increases in a time-dependent, suggesting that ORFV can replicate in breast cancer cells and reach a peak at 72 h. All three types of breast cancer cells have similar results, indicating that ORFV has no limitation on the type of tumor. We also observed the replication and packaging of virus particles in the cytoplasm through a transmission electron microscope (Supplementary Figure 1).

Different MOI (MOI = 0, 1, 2, 5) of ORFV infected three different breast cancer cells, and cell viability were detected through CCK-8 reagent at different time points. As the infection time increases, ORFV can effectively inhibit cell viability. In addition, ORFV with high MOI (MOI = 5) have the most significant inhibitory effect on cell viability (Figure 1B). And we found that after infection of ORFV with high MOI, the clonal formation ability will be effectively inhibited (Figure 1C). Which indicated the long-term inhibitory effect of ORFV on cell viability. On the contrary, even the high MOI ORFV does not have a growth inhibitory effect on normal cells (Supplementary Figure 2).

In order to investigate the mechanisms of ORFV inhibiting cell growth, we evaluated the changes in the cell cycle progression of breast cancer cells using flow cytometry. We observed that the cell cycle of MDA-MB-231, MCF-7, and 4T1 infected by ORFV will be blocked in the G2/M phase, which means that the mitotic process of cells is inhibited (Figure 2A). Furthermore, the western blot results demonstrated the change of cell cycle after MDA-MB-231 infection with ORFV for 24 h. It can be observed that cyclin B was significantly reduced while CDK2 and cyclin E were slightly decrease. The changes in Rb and p-Rb are not significant (Figure 2B).

The flow cytometry used to detect the effect of ORFV-induced apoptosis in breast cancer cells. The results showed that the number of total apoptotic cells was significantly higher in cells infected with ORFV for 48 h (Figure 2C). ORFV infection of MDA-MB-231 cells for 48 h led to a significant stimulation of caspase-3 and PARP activity. But there is no significant difference in the ratio of Bcl-2/bax protein expression (Figure 2D). We speculate that ORFV may play a role in the downstream of apoptosis.

We have verified that ORFV treatment does not produce serious toxicity in vivo experiments on nude mice (Supplementary Figure 3). Some studies have reported that OVs can regulate the TME and activate the innate and adaptive anti-tumor immune response. Then we investigated the anti-tumor effect of ORFV in 4T1 tumor-bearing mice with normal immunity. The animal treatment protocol is shown in Figure 3A. We found that tumor growth can be significantly repressed after intratumoral injection of ORFV (Figures 3B,C). The result was confirmed by immunofluorescence. As shown in Figure 3D, a significantly reduced in the number of Ki67-positive cells in the ORFV group (intratumoral injection of ORFV) in comparison with the Mock group. The c-caspase3-positive cells in the ORFV group were significantly higher than those in the Mock group, which proved that ORFV exerted oncolytic effects by inducing apoptosis (Figure 3E). As expected, western blot results showed that expression of c-Caspase-3 increased in tumor tissues after ORFV injection. In addition, we also detected the expression of c-Caspase-1, which is associated with apoptosis and inflammation. Its expression was significantly increased in ORFV-injected tumor tissues, suggesting that ORFV may cause an inflammatory response at the tumor site and change the tumor microenvironment (Figure 3F).

To determine whether ORFV could affect the immune system, we assessed the proportion of immune cells in the spleen. HE-staining results revealed that the phenotype of lymphocytes in the Mock group was normal. As compared with the Mock group, the spleen demarcation between the red pulp and white pulp was not clear in ORFV group. And we observed a significant increase in neutrophils and the appearance of secondary follicles (Figure 4A yellow triangular markers). Due to immune stimulation by antigens, primary lymphoid follicles, with concentrated chromosomes, sparse cytoplasm, and compact arrangement, will become secondary lymphoid follicles with larger cell size, with 1–3 small nucleoli distributed in the nuclear membrane edge. Increased numbers of neutrophils and secondary lymphoid follicles indicate activation of the body’s immune system. The T cell antigen receptor-CD3 complex plays an important role in the recognition and response to antigens. The IHC results showed that more CD3-positive cells appeared in the spleen of the ORFV group, which means that the body’s T cell immunity was activated (Figure 4B). Meanwhile, the intratumoral injection of ORFV changes the TME by up-regulating the chemokines CCL3 and CCL19, thereby activates the immune response (Figure 4C).

The screening protocol is shown in Figure 5A. Specific primers were used for PCR, and the PCR products were recovered and purified and subjected to next-generation sequencing (Figure 5B). Results of candidate kinase screening are subjected to protein- protein interaction (PPI) network display, Kyoto Encyclopedia of Genes and Genomes Ontology (KO) pathway analysis and Gene Ontology (GO) analysis (Supplementary Data 1 and Supplementary Figure 4). We first screen the original sequencing data, sort the read values of the data, and finally screen out candidate kinases with higher read value differences (Figure 5C).

Then we designed some corresponding siRNAs for verification. After siRNA transfected MCF-7 cells for 24 h, add MOI = 1 ORFV with green fluorescent, continue to culture for 24 h. Set untreated MCF-7 cells as control. The proportion of cells infected by the virus was calculated through detecting GFP by flow cytometry. Flow cytometry results show that siRNA silencing the expression of SGK3, ULK3, ERBB4, STK17B can reduce the infection efficiency of ORFV-GFP virus (Figure 5D), while siRNA silencing the expression of MST1R, PAK4, ERBB3 can increase the infection of ORFV-GFP virus efficient (Figure 5E).

PF-3758309 is the only ATP-competitive PAK inhibitor to have reached clinical trials (Murray et al., 2010). In this section, we investigate whether PAK4 inhibition with siRNA or PF-3758309 could enhance the oncolytic effects of ORFV. According to the grouping, cell samples were collected at 24, 48, and 72 h after ORFV treatment. Through reverse transcription and fluorescent quantitative PCR, it can be found that the mRNA expression level of PAK4 was down-regulated in both the siRNA group and the combination group after siRNA treatment (Supplementary Figure 5). We can also observe that there was no statistical difference in the expression of the viral core gene ORFV035 between the siPAK4 group and the ORFV group at the same time point after treatment (Supplementary Data 2, 3). Among them, Combo group showed that PF-3758309 is beneficial to the replication of ORFV (Figure 6A). The CCK-8 results showed that the Combo group not only showed a significant growth inhibitory effect on MCF-7, but also showed a stronger inhibitory ability compared to the single treatment group (Figure 6B).

The results of flow cytometry detection of cell apoptosis showed that transfection of siPAK4 itself did not cause apoptosis of host cells, and treatment of MCF-7 with 20 nM PF-3758309 for 48 h could induce partial cell apoptosis. The virus itself infects MCF-7 cells for 48 h to induce apoptosis in some cells, and the other two groups of combination groups have a more obvious effect of inducing cell apoptosis than the virus alone (Figure 6C). We designed a series of combined treatments and determine the therapeutic synergism using the Chou-Talalay method. The CIs of all treatment groups were < 1, suggested that PF-3758309 act synergistically with ORFV to enhance the anti-tumor effect in breast cancer (Supplementary Table 1 and Supplementary Figure 6).

We also conducted cell scratch test and transwell invasion assay. Although all three treatment groups can inhibit breast cancer cell migration, there is no significant difference between combo group and single treatment with ORFV or PF-3758309 (Supplementary Figure 7). The result in the transwell invasion assay showed that all three treatment groups can inhibit the ability of breast cancer cell invasion, and the combo group showed the most significant inhibitory ability. The above results indicate that PAK4 inhibitor can enhance the anti-tumor effect of ORFV on breast cancer (Figure 6D).

The animal treatment protocol is shown in Figure 7A. Drawing the tumor growth curve (Figure 7B), it can be observed that the tumors of the mice in the ORFV group and PAK4i group grew slower than the Mock group, and the Combo group showed significant tumor growth inhibition ability. When the tumor volume of the mice in the Mock group reached 500 mm3, the tumor sites of the mice were sampled and compared and analyzed. The tumor sizes of the mice in the three treatment groups were significantly different from those in the Mock group (Figure 7C). The Combo group showed the strongest anti-tumor effect.

Tumors with a high Ki67 positive rate are generally considered to grow faster and have a poor prognosis. Through immunofluorescence detection of tumors in mice, it can be observed that the number of Ki67-positive cells in the tumor site of the Combo group is the least (Figure 7D). The c-Caspase-3 protein is generally considered to be a marker protein for cell apoptosis. Through IHC detection of tumors in mice, it can be observed that the c-Caspase-3 positive cells in the tumor site of the Combo group are much higher than those of the Mock group (Figure 7E).