**2. Results and Discussion**

#### *2.1. PN Suppressed Colorectal Cancer Cell Progress*

To investigate the anti-proliferative e ffects of PN on mutated KRAS-driven CRC, KRAS mutated SW480 and HCT116 cells as well as KRAS-wild type (WT) HT29 and WiDr were evaluated. Cells were treated with PN at 0, 0.1, 0.5, 1, 2, or 4 μg/mL and incubated for 48 or 72 h in triplicate. Cetuximab which has been used to treat mutated KRAS-driven CRCs was used as a control. Figure 1A summarized the proliferation profiles. PN inhibited cell growth in a concentration dependent manner in mutated KRAS-driven cell lines at 48 and 72 h. At 72 h, SW480 viability was decreased to 41.6 ± 1.0% when treated with PN at 2 μg/mL and to 26.0 ± 4.5% when treated with 4 μg/mL PN. When compared to control drug of Cetuximab, PN showed higher suppressive e ffect from 2 μg/mL concentration. Data also indicated that KRAS mutated cell lines responded more sensitively on PN treatment compared to KRAS wild type cells of HT-29 and WiDr. In KRAS wild type cells, PN treatment did not show statistical decrease on proliferation at 48 h. The IC50 values of KRAS mutated cells at 72 h are 1.74 and 2.78 μg/mL on SW480 and HCT116 cells, whereas they are 4.14 and 4.46 μg/mL on KRAS WT cells of HT29 and

WiDr cells. IC50 values of KRAS mutated cells are relatively low compared to other natural extract such as pogostone (HCT116: 18.7 ± 1.93 μg/mL) [24].

**Figure 1.** (**A**) PN suppressive effect on KRAS-mutated colorectal cancer cells of SW480 (KRASG12V) and HCT116 (KRASG13D), and KRAS-wild types CRCs of HCT116 and WiDr. Cells were treated for 48 and 72 h under PN treatment (0, 0.1, 0.5, 1, 2 and 4 μg/mL) and Cetuximab (30 μg/mL). IC50 values are calculated. Results are presented as means ± S.D. of three independent experiments. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001. (**B**) Representative colorectal cancer cell images under PN treatment (0, 0.5, 1 and 4 μg/mL). Blue represents the DAPI-stained cell nuclei, and the propidium iodide-stained dead cells are red. Scale bar = 500 μm. DAPI, 4-,6-diamidino-2-phenylindole; PI, propidium iodide.

Cells were stained with DAPI and PI for live and dead cell detection. Cell nuclei were stained with DAPI (blue) which represent all cells, and dead cells were stained with PI (red). Most cells in control group were stained with DAPI and little PI staining after 48 h of cultivation indicating most cells are alive (Figure 1B). In contrast, cells treated with PN began to appear dead PI stained cells from 0.5 μg/mL.

By assessing clonogenic potentials, we evaluated if PN inhibited tumorogeneity. As shown in Figure 2, PN dose-dependently suppressed colony formation by the KRAS mutated cell lines SW480, and HCT116. For example, PN treatment at 1 μg/mL suppressed SW480 cell colony formation to 26%. Surprisingly, almost no colony formation was observed by KRAS mutated CRCs treated with 4 μg/mL of PN. In contrast, HT29 and WiDr colonies are still visible even at the highest concentration. These findings show that PN sensitively suppress proliferation rate and tumor forming abilities in the long-term aspects of mutated KRAS-driven CRC cells. Further mechanistic and molecular studies were performed with SW480 cells at PN concentrations ranging from 0.1 to 4 μg/mL.

**Figure 2.** (**A**) Clonogenic potential after the PN treatment (0, 1 and 4 μg/mL) on KRAS-mutated colorectal cell lines of SW480 (KRASG12V) and HCT116 (KRASG13D) and KRAS-WT cells of HT29 and WiDr. (**B**) Relative colony numbers upon the PN treatment were calculated compared to the control. Results are presented as means ± S.D. of three independent experiments. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

Seeds of *Pharbitis nil* contains several bioactive compounds such as polysaccharides [16], pharbitin (resin glycosides) [25], anthocyanins, diterpenoids [26], triterpene saponins [27], pharbilignan C [17], neolignane, and monoterpene glycosides [28]. Pharbilignan C has been reported to induce human breast cancer cell apoptosis via the mitochondria-medicated intrinsic pathway [17], and lignans exhibited anti-inflammatory and anti-cancer activities [29]. Thus, the observed suppressive e ffects of PN on CRC cell-lines may have been due to the presence of these functional compounds.

#### *2.2. PN Induced Apoptosis and Cell Cycle Arrest in the G2*/*M Phase*

To investigate the mechanism of cell death in mutated KRAS-driven human colorectal cell lines induced by PN, we analyzed apoptosis and cell cycles. For detecting apoptotic cells, Annexin V-FITC and PI staining was used. Annexin V binds specifically to a phosphatidylserine residue that is externalized by cells undergoing apoptosis. As seen in Figure 3, PN substantially increased the percentage of apoptotic cells in a concentration-dependent manner. Early (Annexin V FITC+/PI−) and late (Annexin V FITC+/PI<sup>+</sup>) apoptosis counts were increased up to 28.6 ± 1.4 % and 40.2 ± 2.7 % at 2 or 4 μg/mL PN treatment for 48 h, respectively. These observations sugges<sup>t</sup> PN inhibits CRC proliferation by inducing apoptosis.

**Figure 3.** (**A**) Flow cytometric apoptosis images PN treatment using Annexin V-FITC/PI staining. SW480 cells were stained with Annexin V-FITC/PI upon PN treatment (0, 0.1, 0.5, 1, 2, and 4 μg/mL). (**B**) The percentage of apoptotic cells upon the PN treatment were compared. Asterisks (\*) indicate statistical differences compared to untreated control (n = 3). \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001; Student's *t* test. Statistical differences among the experimental groups were confirmed via one-way ANOVA test. FITC, fluorescein isothiocyanate; PI, propidium iodide.

Cell cycle is progressed into four phases: gap1 (G1), DNA synthesis (S), gap2 (G2) and mitosis (M). Arresting cancer cells at a certain stage are often viewed as therapeutic targets [30]. The cell cycle after exposing SW480 cells at concentrations of 0.5–4 μg/mL of PN was investigated. As seen in Figure 4A, cell cycle analysis showed PN concentration-dependently induced accumulation in the G2/M phase and a concomitant reduction in the G0/G1 phase. For example, the proportions of cells in the G2/M phase after treatment with 1 or 4 μg/mL of PN were 28.5 ± 0.3% and 41.2 ± 6.7%, respectively, which demonstrated PN inhibited cell growth by arresting cells in the G2/M phase. Then, we investigated expression levels of the cdc2 and cyclin B1, which are proteins for cell cycle progression from G2/M to G0/G1. Indeed, western blot images revealed that PN concentration-dependently reduced the levels of both proteins (Figure 4B).

PN was found to arrest cells in the G2/M phase to reduce the number of cells in the G0/G1 phase (Figure 4), therefore less cells enter into mitosis. This is interesting in that most of the cell cycle arrests caused by natural extracts reported to date have occurred in the G0/G1 or G2/M stages. [21,31,32]. For example, pogostone induced G0/G1 arrest in a KRAS mutated HCT116 cell-line [24].

**Figure 4.** (**A**) Cell cycle analysis upon PN treatment (0, 0.5, 1, 2, and 4 μg/mL). Images represent the SW480 cell response under PN treatment. Percentages in each cell cycle phase (G0/G1, S, G2/M) are calculated and G2/M phase are increased as concentration dependent manner. Asterisks indicates statistical differences compared to untreated control through Student's *t*-test (\* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001). (**B**) Expression of G2/M phase related proteins of cdc2 and cyclin B1 upon PN treatment. HSP90 was used as a loading control and the protein intensity was analyzed in triplicate. The ratios of cdc2/HSP90 and cyclin B1/HSP90 were calculated and analyzed by Student *t*-test \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

#### *2.3. Inhibition of the AKT*/*mTOR Pathway Enhanced PN-Induced Cell Death*

Mutated KRAS-driven CRC signals are through the ERK/MEK or AKT/mTOR pathways [3,4]. During mutated KRAS-driven CRC progression, mutant RAS constitutively activates ERK/MEK or AKT/mTOR phosphorylation, and as a result promotes cancer cell proliferation. To determine whether SW480 growth inhibition by PN is regulated by RAS/ERK or AKT/mTOR pathways, cells are treated with PN for 48 h and protein levels of KRAS, p-ERK, p-AKT, and p-mTOR were assessed by western blot. As shown in Figure 5, PN significantly and concentration-dependently suppressed p-AKT and p-mTOR phosphorylation. For example, the expression of p-AKT was reduced to 0.31 in PN 4 μg/mL treated cells as compared with 1.0 in the non-treated control. The level of KRAS and ERK phosphorylation was similar in all groups as observed in Figure 5. These results demonstrated that PN suppressed CRC progression predominantly via the AKT/mTOR pathway. The AKT/mTOR and ERK/MEK pathways are known to be major regulatory signaling pathways that relate to CRC cell proliferation, metabolism, and survival [2,4,33]. Our observations sugges<sup>t</sup> that the antitumor effect of PN is caused by reductions in the phosphorylations of AKT and mTOR.


**Figure 5.** (**A**) Representative western blot images of KRAS, HSP90, phospho-p42/<sup>44</sup> MAPK (phospho-ERK1/2), phospho-AKT, phospho-mTOR protein expression in SW480 cells treated with the compound PN (0, 0.5, 1, 2 and 4 μg/mL). (**B**) The ratios of KRAS/HSP90, phospho-ERK1/2/HSP90, and phospho-AKT/GAPDH, phospho-mTOR/GAPDH were calculated and compared to the control. (N = 3; mean SEM, \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001; Student's *t* test).

#### *2.4. PN Restores Muscle Cell Function during Cancer Progression*

During cancer progression, skeletal muscles are weakened which causes progressive functional impairment, called cachexia [21–23]. Accordingly, it is important aspects that agents targeting cancers do not adversely affect muscle cells. Thus, we examined the proliferation and function of muscle cells of myoblast, treated with different concentrations of PN (0, 0.1, 0.5, 1, 2, or 4 μg/mL) for 48 or 72 h. In Figure 6A, PN did not inhibit cell proliferation up to 2 μg/mL concentrations. After 72 h of treatment with 2 or 4 μg/mL of PN, cell viabilities remained at 97% and 61% of non-treated control

levels. Next, we examined if PN affected muscle functions of myotube formation by inducing myogenic differentiation. Phenotypes were detected by myosin heavy chain (MyHC) immunostaining and fusion indices were calculated. Surprisingly, PN treatment did not inhibit myogenic differentiation as determined by MyHC staining and fusion indices (Figure 6B,C). In fact, fusion indices even improved at all PN treated groups. Our data sugges<sup>t</sup> that PN has a beneficial effect on muscle cell function.

**Figure 6.** (**A**) C2C12 cells were treated with different PN concentrations (0, 0.1, 0.5, 1, 2 and 4 μg/mL) for 48 and 72 h. Results are presented as means ± S.D. of 3 independent experiments, \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001. (**B**) Immunofluorescence microscopy for the expression of the myogenic markers Myosin heavy Chain (MyHC) and DAPI. Myogenesis was induced in differentiation media and treated for 5 days with different PN concentration (0, 0.5 and 1 μg/mL). (**C**) Fusion indexes were calculated as the % of the nuclei inside myotubes compared to the total number of nuclei. (N= 3 independent experiments; mean SEM, \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001; Student's *t* test). Between experimental groups, statistical difference was found via one-way ANOVA (### *p* < 0.001). (**D**) Representative images of myotube formation on cancer conditioned media upon PN treatment (0, 0.1, 0.5 and 1 μg /mL). Myotube was detected with MyHC (green) immunostaining and nuclear counterstained DAPI (blue). Scale bar = 200 μm. (**E**) Fusion index were calculated as the % of the nuclei inside myotubes compared to the total number of nuclei. (Data were from three independent experiments; mean SEM, \* *p* < 0.05; Student's *t* test).

Because PN did not impair and rather supported C2C12 myogenic differentiation into myotube formation, we further investigated if PN could attenuate conditioned media (CM)-mediated muscle function impairment. In Figure 6D,E, reduced myotube numbers and MyHC expression were observed on PN-untreated control, which indicated cancer environment significantly impaired muscle cell function. However, PN-treatment rescued C2C12 muscle cell dysfunctions. In detail, increased myotube formation and MyHC expression compared to the SW480 control CM were observed (Figure 6D,E).

During cancer progression, cancer-associated cachexia is common [21], and one of the problems posed by anti-cancer drugs is that they impair normal cell proliferation and function. Therefore, the maintenance of muscle cell proliferation and myogenic differentiation are important aspects at the cancer therapeutic development. Interestingly, we found PN did not impair muscle cell proliferation, but actually restored muscle cell function.
