Inhibition of Polo-like kinase 1 (PLK1) triggers cell apoptosis via ROS-caused mitochondrial dysfunction in colorectal carcinoma

Colorectal cancer (CRC) is one of the most frequently diagnosed cancers. Polo-like kinase 1 (PLK1), a member of the serine/threonine kinase PLK family, is the most investigated and essential in the regulation of cell cycle progression, including chromosome segregation, centrosome maturation and cytokinesis. However, the nonmitotic role of PLK1 in CRC is poorly understood. In this study, we explored the tumorigenic effects of PLK1 and its potential as a therapeutic target in CRC. GEPIA database and immunohistochemistry analysis were performed to evaluate the abnormal expression of PLK1 in CRC patients. MTT assay, colony formation and transwell assay were performed to assess cell viability, colony formation ability and migration ability after inhibiting PLK1 by RNAi or the small molecule inhibitor BI6727. Cell apoptosis, mitochondrial membrane potential (MMP) and ROS levels were evaluated by flow cytometry. Bioluminescence imaging was performed to evaluate the impact of PLK1 on CRC cell survival in a preclinical model. Finally, xenograft tumor model was established to study the effect of PLK1 inhibition on tumor growth. First, immunohistochemistry analysis revealed the significant accumulation of PLK1 in patient-derived CRC tissues compared with adjacent healthy tissues. Furthermore, PLK1 inhibition genetically or pharmacologically significantly reduced cell viability, migration and colony formation, and triggered apoptosis of CRC cells. Additionally, we found that PLK1 inhibition elevated cellular reactive oxygen species (ROS) accumulation and decreased the Bcl2/Bax ratio, which led to mitochondrial dysfunction and the release of Cytochrome c, a key process in initiating cell apoptosis. These data provide new insights into the pathogenesis of CRC and support the potential value of PLK1 as an appealing target for CRC treatment. Overall, the underlying mechanism of inhibiting PLK1-induced apoptosis indicates that the PLK1 inhibitor BI6727 may be a novel potential therapeutic strategy in the treatment of CRC.


Introduction
Colorectal cancer (CRC) is one of the most common malignancies with a high mortality rate in the world (Liu et al. 2017a). Because of the absence of specific and distinct symptoms, most CRC patients are diagnosed in the advanced unresectable stage, accompanied by poor prognosis. To date, the complex molecular mechanisms of CRC pathogenesis are only partially known, which hinders the treatment of CRC. Consequently, understanding the underlying molecular mechanisms of CRC occurrence and development is of great significance for the identification of new therapeutic targets and prognosis of CRC.
Polo-like kinases (PLKs) are a family of five highly conserved serine/threonine protein kinases that are involved in the regulation of cell cycle progression and DNA damage response (Weerdt and Medema 2006). Polo-like kinase 1 (PLK1), a member of the PLK families, is characterized by an N-terminal serine/threonine kinase domain (KD) and a C-terminal polo-box domain (PBD), which mediates substrate recognition and specific subcellular localization, and regulates kinase activity (Carcer et al. 2011;Jang et al. 2002). The functions of PLK1 are traditionally linked to the regulation of the cell cycle and stress response. In recent years, more functions of PLK1 have been proposed than just their traditional roles in various aspects of mitosis (Raab et al. 2021). Increasing evidences have demonstrated that PLK1 contributes to the regulation of DNA replication (Yim and Erikson 2009), mTOR signaling and autophagy, apoptosis (Ruf et al. 2017), and the epithelial-to-mesenchymal transition (Fu and Wen 2017). The new roles of PLK1 broaden the therapeutic potential of targeting PLK1.
Mitochondrial apoptosis is the most commonly deregulated form of cell death and plays a key role in cancer development . Different apoptotic stimuli, such as mitochondrial DNA damage and reactive oxygen and nitrogen species, generally result in mitochondrial outer membrane permeabilization, which is regulated largely by members of the Bcl2 family (Circu and Aw 2008;Lopez and Tait 2015). Subsequently, the release of several proapoptotic factors from the mitochondrial intermembrane into the cytosol follows, notably Cytochrome c and Smac/Omi, leading to caspase activation and cell death through the parallel cleavage of hundreds of different substrates (Li et al. 2016).
Numerous studies have shown that PLK1 is overexpressed and involved in the regulation of tumor occurrence and malignant biological behavior of various cancers including colorectal cancer (Klauck et al. 2018), melanoma (Chen and Villanueva 2015), prostate cancer (Liu et al. 2011), non-small cell lung cancer , and breast cancer (Weichert et al. 2005). Upregulated PLK1 signaling is accompanied by deregulation of cell cycle-related pathways and poor prognosis in CRC patients, indicating its potential as an attractive target in CRC treatment (Yu et al. 2021). However, the non-mitosis role of PLK1 in CRC largely remains unclear. In particular, little is known about the function of PLK1 in the regulation of mitochondrial apoptosis.
In this study, we explored the relationship between PLK1 overexpression and tumor biological behavior in CRC. In addition, we explored the underlying mechanism of PLK1 inhibition-induced apoptosis. Our data suggested that PLK1 inhibition significantly reduced cell viability, migration, and colony formation ability and induced apoptosis of CRC cells, which was triggered by ROS-mediated mitochondrial dysfunction. These data provide new insights into the pathogenesis of CRC and support the potential value of PLK1 as an appealing target for CRC treatment.

Cell lines and culture
Both normal colon (FHC) and colon cancer (HCT116 and DLD-1) cell lines were purchased from ATCC. FHC and HCT116 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% FBS and 1% penicillin/ streptomycin. DLD-1 cells were cultured in RPMI 1640 medium-containing 10% FBS and 1% penicillin/streptomycin. All cultures were maintained at 37 °C with a 5% CO 2 atmosphere in a humidified incubator.

Gene expression profiling interactive analysis (GEPIA)
The online database GEPIA (http:// gepia. cancer-pku. cn/ index. html) was used to analyze the differential expression of PLK1 in CRC tumor and normal tissue datasets. GEPIA is a newly developed interactive web server for analyzing the RNA sequencing expression data of 9,736 tumor and 8,587 normal samples from the TCGA and GTEx projects . Survival curves, including OS and DFS, were obtained based on gene expression with the log-rank test and the Mantel-Cox test via the GEPIA database.

Gene expression profiling interactive analysis (UALCAN)
UALCAN (http:// ualcan. path. uab. edu/ analy sis. html), a comprehensive web resource, provides analyses based on RNA sequence and clinical data relative to 31 cancer types from The Cancer Genome Atlas (TCGA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC) databases (Chandrashekar et al. 2017). The correlation between PLK1 mRNA expression and CRC TNM stages was evaluated using the UALCAN database.

Lentivirus production and transfection
The lentiviral control or PLK1 shRNA plasmid was transfected into HEK293T cells to produce lentiviral particles. Then, HCT116 and DLD-1 cell lines were established with stable knockdown of PLK1 or its scrambled controls. Single colonies were obtained after 2 weeks of selection with 1 μg/mL puromycin. The negative control siRNA and PLK1 siRNA were transiently transfected into HCT116 or DLD-1 cells using Lipofectamine 2000 transfection reagent (Invitrogen, USA) according to the manufacturer's instructions.

Western blot analysis
Total protein was extracted using RIPA lysis buffer, and equal amounts of protein were separated by SDS-PAGE (6-15%) and transferred onto polyvinylidene fluoride (PVDF) membranes, which were blocked with 3% bovine serum albumin (BSA) for 60 min. The membranes were incubated with specific primary antibodies overnight at 4 °C. Specific primary antibodies against PLK1 (#4513) and Ki67 (ab16667) were purchased from Cell Signaling Technology. Antibodies against COX4 (sc-376731), Bcl2 (sc-7382), and Bax (sc-7480) were obtained from Santa Cruz Biotechnology. Antibody against Cytochrome c (AF0146) was obtained from Affinity. The protein levels were normalized to β-actin (EASYBIO).

RNA extraction and RT-qPCR
Total RNA was extracted with TRIzol reagent (Ambion, Lot No. 317110). Two micrograms of RNA was used for reverse transcription to cDNA by reverse transcriptase (YEASEN, catalog No. 11141ES60) following the manufacturer's instructions. cDNA (500 ng) was used for RT-qPCR with SuperReal PreMix Plus (SYBR Green) reagent (TIANGEN,. The reaction parameters were as follows: 95 °C for 15 min, followed by 45 cycles of amplification with three steps: denaturation at 95 °C for 10 s, annealing at 55 °C for 20 s, and extension at 72 °C for 30 s. The primer sequences were as follows: human β-actin forward CAC CAT TGG CAA TGA GCG GTTC; human β-actin reverse AGG TCT TTG CGG ATG TCC ACGT; human PLK1 forward CAC CAG CAC GTC GTA GGA TTC; human PLK1 reverse CCG TAG GTA GTA TCG GGC CTC.

Cell viability assay
HCT116 or DLD-1 cells were seeded into 96-well plates at a density of 4 × 10 3 cells/well. After 24 h, cells were treated with the indicated doses of BI6727 (Selleck, S2235) or transiently transfected with PLK1 siRNA. Cell viability was determined by MTT or CCK-8 assay.

Colony formation assay
The effect of PLK1 on the colony formation ability of CRC cells was evaluated by the colony numbers. HCT116 or DLD-1 cells were seeded into 6-well plates at a density of 2-3 × 10 3 cells/well for 24 h. Then, the cells were treated with the indicated doses of BI6727 for 48 h. After that, the medium was replaced with fresh medium-containing 2% FBS, and the cells were cultured for 14 days with media changes every third day. Treated cells were washed with PBS, fixed with 10% formalin for 10 min and stained with 0.05% crystal violet for 30 min.

Cell migration assay
The bottom chambers were filled with fresh medium-containing 10% FBS. Then, 5-8 × 10 4 cells/well were seeded into the upper chambers in serum-free medium with or without BI6727. After incubation at 37 °C for 48 h, the cells on the lower surface of the membrane were washed with PBS, fixed with 10% formalin for 10 min and stained with 0.05% crystal violet for 30 min.

Cell apoptosis assay
Apoptosis assays were carried out with an Annexin V-APC/7-AAD apoptosis detection kit (MultiSciences, catalog 70-AP105-100) or Annexin V-FITC/PI apoptosis detection kit (MultiSciences, catalog 70-AP101-100) according to the manufacturer's protocol. After BI6727 treatment or PLK1 knockdown, HCT116 or DLD-1 cells were collected and resuspended in binding buffer. Sequentially, the cells were labeled and analyzed by flow cytometry (BD LSR-Fortessa). Apoptosis was quantified as the Annexin V-positive cell numbers.

Intracellular ROS detection
Intracellular ROS production was evaluated using an ROS assay kit (Solarbio, CA1410). Treated HCT116 or DLD-1 cells were collected and co-incubated with DCFH-DA for 30 min at 37 °C. After washing twice with PBS, the cells were immediately quantitated by flow cytometry at an emission wavelength of 525 nm and an excitation wavelength of 488 nm.

Detection of mitochondrial membrane potential (MMP)
The mitochondrial membrane potential (MMP) was determined by the mitochondrial staining kit (JC-1, Multi-Sciences Biotech, catalog 70-MJ101). HCT116 cells were treated with BI6727 for 24 h or transiently transfected with PLK1 siRNA for 48 h, and then, JC-1 was loaded for 20 min. The emissions at 530 nm and 590 nm were analyzed by flow cytometry (BD LSRFortessa) using an excitation wavelength of 488 nm. The ratio of mean fluorescence intensity (JC-1 aggregates/monomer) was calculated to define MMP levels.

Mitochondrial isolation
Following BI6727 treatment or PLK1 siRNA transfection, HCT116 or DLD-1 cells were collected, and mitochondria were isolated using the Cell Mitochondria Isolation Kit (Beyotime). The efficiency of mitochondrial and cytoplasmic fractionation was confirmed by Western blot analysis against the mitochondrial protein COX4. The levels of Bax and Cytochrome c in the mitochondria and cytoplasm were examined by Western blot analysis.

Immunofluorescence staining
HCT116 cells were plated on the glass coverslips in 24-well cell culture plates overnight with PLK1 knockdown or BI6727 treatment. The cells were fixed with 10% formalin for 10 min, permeabilized with 0.2% Triton X-100 for 10 min, and blocked in 10% goat serum for 1 h. For immunostaining, the cells were incubated with primary antibody overnight at 4 °C and then incubated with secondary antibodies conjugated with Alexa Fluor 488 or Alexa Fluor 594 for 1 h. Images were captured with a confocal laser scanning microscope.

Immunohistochemistry (IHC)
HCT116 xenograft tumor tissues were fixed in 4% paraformaldehyde, embedded in paraffin wax, cut into 5 μm sections, and placed onto glass slides. Next, the expressions of PLK1, Bax, and Ki67 in xenograft tumor tissues were analyzed by IHC.

TUNEL assay
Paraffin-embedded tissues from HCT116 xenograft tumors were cut into 4 μm-thick sections and placed onto glass slides. Apoptotic cells were identified using the TUNEL Assay Apoptosis Detection Kit (FITC) (ABSIN, abs50033) according to the manufacturer's instructions. Representative images were captured with a confocal laser scanning microscope.

Bioluminescence imaging of the intravenous tumor model
The luciferase-labeled HCT116 cells were transiently transfected with small interfering RNA (siRNA) targeting PLK1 or negative control siRNA (siNC). Subsequently, equal numbers of HCT116 cells (5 × 10 5 /100 μl PBS) were injected into BALB/c female mice intravenously. Ten minutes prior to imaging, mice were administered 120 mg/kg D-luciferin (PerkinElmer, catalog 122,799) by intraperitoneal injection. Bioluminescent imaging was captured at the indicated times using the Xenogen IVIS Spectrum System, and the total photon flux of chest regions was quantified.

Xenograft tumor model
Female BALB/c nude mice at 6 to 8 weeks of age were maintained under specific pathogen-free conditions. Equal numbers (3.5 × 10 6 ) of HCT116 cells were resuspended in 0.1 mL PBS and then injected subcutaneously into the left and right flanks of mice. When the tumor volumes reached approximately 400 mm 3 (volume = length × width 2 /2), the mice were randomly divided into two groups and treated with BI6727 (10 mg/kg body weight). Tumor volume was monitored every 2 days. After 22 days, tumors were collected. All experiments involving mice were approved by the Animal Care Committee of Nankai University, Tianjin, China.

Statistical analysis
All data are represented as the mean ± SEM. In vitro and vivo studies, statistical analyses were performed using a multiple t test to compare differences among multiple groups. Student's t test of column analysis was applied to compare differences between two groups. P < 0.05 was considered statistically significant.

Abnormal PLK1 expression was identified in CRC
Overexpression of PLK1 has been demonstrated in a wide range of human malignancies, including ovarian, breast, lung, and colon cancers. In accordance with the previous studies, Gene Expression Profiling Interactive Analysis (GEPIA) (Tang et al. 2017) showed that PLK1 was abnormally expressed in CRC tissues compared with normal tissues (Fig. 1A). Analysis with the UALCAN databases further indicated that the mRNA expression level of PLK1 in colorectal cancer tissues of different TNM stages was higher than that in normal colorectal tissue (Fig. 1B). Moreover, RT-qPCR analysis showed that PLK1 mRNA expression was significantly increased in HCT116 cells compared with the normal colonic epithelial cell line FHC (Fig. 1C). Western blot analysis also demonstrated a consistent trend in which PLK1 protein levels were notably higher in the CRC cell lines than in the FHC cell line (Fig. 1D). In addition, immunohistochemical staining analysis further demonstrated that the expression of PLK1 was higher in CRC patient tissues than in adjacent normal tissues (Fig. 1E). These results indicated that dysfunction of PLK1 signaling pathway might promote tumorigenesis in CRC, and targeting PLK1 signaling could be a potential therapeutic strategy against CRC.

Depletion of PLK1 reduced proliferation and induced apoptotic cell death of CRC cells
To determine the effects of PLK1 on CRC tumor growth, HCT116 and DLD-1 cell lines with stable PLK1 knockdown (shPLK1) or a negative control vector (shNC) were constructed by lentiviral transfection. The knockdown efficiency was evaluated by RT-qPCR and western blot analyses ( Fig. 2A and 2B). Colony formation and Transwell assays were performed to assess colony formation ability and migration ability, respectively ( Fig. 2C and 2D). Subsequently, HCT116 and DLD-1 cells were transiently transfected with siRNA targeting PLK1 (mixed siPLK1#1 and siPLK1#2) or N.C. Cell apoptosis was assessed using flow cytometry, and cell proliferation was analyzed by the cell counting kit-8 (CCK-8) assay ( Fig. 2E and 2F). These results suggested that PLK1 depletion inhibited cell proliferation, colony formation and cell migration, and obviously induced apoptosis of CRC cells compared with the control. To further investigate the impact of PLK1 on CRC cell survival in a preclinical model, luciferase-labeled HCT116 cells were transiently transfected with PLK1 siRNA or control siRNA and then intravenously injected into BALB/c mice. As indicated by bioluminescence intensity, PLK1-depleted cells showed less lung colonization than control siRNAtransfected cells at 6 and 9 h after cell injection (Fig. 2G). These results indicated that depletion of PLK1 impeded tumorigenesis both in vitro and in vivo.

Pharmacological inhibition of PLK1 by BI6727 decreased cell proliferation and triggered apoptosis in vitro
BI6727 (volasertib), a potent small molecular inhibitor of PLK1 that targets the kinase domain, is currently being tested in clinical trials for multiple cancer treatments (Liu 2015). To further characterize the role of PLK1 in the carcinogenesis of CRC, HCT116 and DLD-1 cells were treated with BI6727 at gradient concentrations. The MTT assay showed that BI6727 decreased cell viability in a dose-dependent manner with a half maximal inhibitory concentration (IC50) of 27.13 nM in HCT116 cells and 67.35 nM in DLD-1 cells (Fig. 3B). Furthermore, BI6727 caused a decrease in cell numbers and degeneration of CRC cell morphology (Fig. 3A) and significantly induced CRC cell apoptosis (Fig. 3C). Colony formation assays showed that BI6727 significantly inhibited the colony formation ability of CRC cells in a dose-dependent manner (Fig. 3D). In addition, Transwell assays revealed that CRC cells treated with BI6727 showed decreased migratory ability compared with control cells (Fig. 3E). These results suggested that pharmacological inhibition of PLK1 contributed to decreased proliferation and migration ability and the induction of apoptosis in CRC cells, which were consistent with the effects of knocking down PLK1. The knockdown efficiency of PLK1 by specific shRNA in HCT116 and DLD-1 cells was determined by RT-qPCR and Western blot assays, respectively. C Representative images of the colony formation assay and quantification of colony numbers in HCT116 and DLD-1 cells transfected with control or PLK1 shRNA. D Cell migration was analyzed in HCT116 and DLD-1 cells transfected with control or PLK1 shRNA through Transwell assays without Matrigel, and the extent of migration was quantified by relative migrated cell numbers. E The apoptotic cell numbers (Annexin V-positive staining) in HCT116 cells were assessed by flow cytometry after PLK1 knockdown by shRNA or siRNA. F HCT116 and DLD-1 cells were transiently transfected with PLK1 siRNA or control siRNA. CCK-8 assays were performed to evaluate cell viability at 24 h, 48 h, and 72 h post-transfection. G Luciferaselabeled HCT116 cells were transfected with PLK1 siRNA or control siRNA for 48 h, harvested, and intravenously injected into BALB/c mice (n = 5). Bioluminescence intensity was monitored at 0 h, 3 h, 6 h, and 9 h post-injection. Student's t test, *P < 0.05, **P < 0.01, ***P < 0.001

Targeting PLK1 caused apoptotic cell death through downregulation of Bcl2/Bax expression in CRC cells
Our studies have suggested that PLK1 inhibition could markedly induce the apoptosis of CRC cells. The proapoptotic protein Bax and anti-apoptotic protein Bcl2 are two important members of the Bcl2 family involved in apoptosis. Moreover, studies have demonstrated that the Bcl2/Bax ratio is critical for the integrity of mitochondria and caspase-3 activation during apoptosis (Tait and Green 2010;Quan et al. 2020). To determine the mechanism of cell apoptosis induced by PLK1 inhibition, HCT116 and DLD-1 cells were transiently transfected with siRNAs targeting PLK1 (mixed siPLK1#1 and siPLK1#2) or N.C. Western blot analysis showed that PLK1 knockdown significantly enhanced the expressions of apoptosis-related proteins, including cleaved PARP and cleaved caspase-3, and resulted in a strong downregulation of Bcl2, decreasing the Bcl2/Bax ratio (Fig. 4A). A consistent tendency in related proteins was also demonstrated in HCT116 and DLD-1 cells treated with BI6727 (Fig. 4B). Meanwhile, GEPIA survival analysis showed that low Bax expression was associated with poor overall survival (OS) (p < 0.05) in CRC patients (Fig. 4C). High Bcl2 expression was correlated with poor OS in CRC patients, although the difference was not significant (p = 0.5) (Fig. 4D). Furthermore, we analyzed the expression of Bax in CRC patient tissues. Compared with the adjacent tissues, the expression of Bax in CRC tissues was significantly lower, which was associated with poor OS in CRC patients (Fig. 4E). These results demonstrated that PLK1 inhibition prevented antiapoptotic Bcl2 expression and reduced the Bcl-2/Bax ratio, which induced apoptosis of CRC cells.

Apoptosis by PLK1 inhibition was triggered by elevated oxidative stress and impeded mitochondrial function in CRC cells
Reactive oxygen species (ROS), a part of normal cell metabolism, have been proven to induce cell apoptosis once the level of ROS exceeds the elimination ability of its own antioxidant defense systems (Cerutti 1994). Flow cytometry analysis-based DCFH-DA showed that targeting PLK1 significantly elevated ROS accumulation in HCT116 and DLD-1 cells (Fig. 5A), which may directly mediate the occurrence of cell apoptosis. Moreover, accumulating evidences have shown that Bcl2 family members play an essential role in the mitochondrial apoptotic pathway by directly controlling the permeability of the outer mitochondrial membrane (Ma et al. 2020). Considering the decreased Bcl2/ Bax ratio, we first investigated the change in mitochondrial membrane potential (MMP), which is also a critical indicator of apoptosis. JC-1 staining showed that knocking down or inhibiting PLK1 caused the loss of MMP in HCT116 cells (Fig. 5B). It is well known that Bcl2 downregulation and Bax translocation regulate the permeability of the mitochondrial membrane, the release of Cytochrome c (Cyt-c), and the activation of caspase-3 (Murphy et al. 2000;Narita et al. 1998). To test whether a similar mechanism exists in CRC cells, cytosolic and mitochondrial proteins were separated, and adequate separation of the two compartments was evaluated using Cytochrome c oxidase subunit IV (COX4, mitochondrial marker). Western blot analysis revealed that knocking down or inhibiting PLK1 increased the Bax ratio in mitochondria (normalized to cytoplasm) in HCT116 cells and decreased the Cyt-c ratio in mitochondria in HCT116 and DLD-1 cells ( Fig. 5C and 5E), which represented the translocation of Bax from the cytoplasm to mitochondria and the release of Cyt-c from mitochondria to the cytoplasm, respectively. Further immunofluorescence analysis also demonstrated that after knocking down or inhibiting PLK1, Cyt-c expression in mitochondria was decreased in HCT116 cells (Fig. 5D). Taken together, these results indicated that PLK1 inhibition destroyed mitochondrial membrane integrity (including the loss of MMP and Bax translocation), further caused Cyt-c release from mitochondria to the cytoplasm and caspase-3 activation, and finally led to the apoptosis of CRC cells.

Antioxidants rescued mitochondrial dysfunction-mediated apoptotic cell death in CRC cells
To determine whether the PLK1 inhibition-caused reduction in Bcl2 expression and mitochondrial dysfunction were related to excessive ROS generation in CRC cells. HCT116 cells were treated with H 2 O 2 and DMTU  by Western blot assay. β-Actin was used as an internal reference in the cytosolic fraction, and COX4 was used as an internal reference in the mitochondrial fraction. D Immunofluorescence was used to detect the cellular localization and expression of Cyt-c in HCT116 cells. DAPI was used for nuclear staining. Scale bar: 10 μm. E Western blot analysis of Cyt-c in the cytosolic and mitochondrial fractions of DLD-1 cells following transfection with control siRNA or PLK1-targeting siRNA. Student's t test, *P < 0.05, **P < 0.01, ***P < 0.001 The expression of Cyt-c in cytosolic and mitochondrial fractions was measured and quantified. Student's t test, *P < 0.05, ***P < 0.001 (N,N'-dimethylthiourea, an effective ROS scavenger). As expected, 10 mM DMTU partially blocked the H 2 O 2 -induced accumulation of ROS (Fig. 6A) and cleaved PARP expression, upregulated Bcl2 expression (Fig. 6B), and rescued H 2 O 2 -induced apoptosis of HCT116 cells (Fig. 6C). Next, flow cytometry analysis showed that the generation of ROS was partially inhibited by DMTU in BI6727-treated HCT116 cells (Fig. 6D) and accompanied by increased cell viability (Fig. 6E). JC-1 staining showed that DMTU partially protected HCT116 cells from the loss of mitochondrial membrane potential (Fig. 6F). Western blot analysis showed that DMTU upregulated the Bcl2/Bax ratio and decreased the expression of cleaved PARP and the activation of caspase-3 ( Fig. 6G). Meanwhile, DMTU also attenuated the release of Cyt-c caused by PLK1 inhibition in HCT116 cells (Fig. 6H). In addition, similar results were also observed in DLD-1 cells. DMTU enhanced cell viability and rescued cell apoptosis induced by H 2 O 2 or BI6727 ( Supplementary Fig. 1A,  B). These results demonstrated that DMTU might effectively alleviate mitochondrial dysfunction by directly reducing the accumulation of ROS. In summary, these results revealed that PLK1 inhibition increased the accumulation of ROS, thus causing downregulation of the Bcl2/Bax ratio, which was responsible for mitochondrial membrane integrity and subsequent leakage of Cyt-c, and activation of intrinsic apoptosis of CRC cells (Supplementary Fig. 1C).

BI6727 suppressed tumor growth and induced apoptosis in xenograft tumor models
To further assess the effects of PLK1 in vivo, we subcutaneously injected equal amounts of HCT116 cells into BALB/c nude mice. After 14 days, the mice were randomized into groups and subjected to intraperitoneal treatment with PBS or BI6727 (10 mg/kg body weight) once every 3 days following the schedule (Fig. 7A). Compared with the control group, we observed that BI6727 treatment significantly diminished tumor sizes (Fig. 7B), including tumor volume and tumor weight (Fig. 7C, D). In addition, H&E staining showed more necrotic areas in HCT116 xenografts treated with BI6727 than in control xenografts (Fig. 7E, left panel). Immunohistochemical staining analysis demonstrated that BI6727 treatment significantly decreased the expressions of PLK1 and Ki67, a cell proliferation marker (Fig. 7E, middle panel). To elucidate whether BI6727 was able to induce cell apoptosis in vivo, TUNEL assay analysis of xenograft tumor sections was performed and showed a significantly higher number of apoptotic cells in the BI6727 treatment group (Fig. 7F). Next, we randomly selected four pairs of tissues for western blot analysis. As expected, the BI6727 treatment group showed increased levels of cleaved PARP and Bax (Fig. 7G), which was also consistent with upregulated Bax expression in IHC staining (Fig. 7E, right panel). These results indicated that BI6727 also induced cell apoptosis and further suppressed the progression of CRC in vivo.

Discussion
CRC is a common malignant tumor of the digestive system with high morbidity and mortality rates (Nasseri and Langenfeld 2017;Marmol et al. 2017). Although a variety of CRC treatments have been applied in the clinic, such as surgery, radiation therapy, systemic chemotherapy, and targeted therapy, the therapeutic efficacy of these approaches is limited, and severe side effects cannot be neglected after long-term observation (Brenner et al. 2014). Accordingly, the identification of new therapeutic targets is critical to improve the survival rate of CRC patients. PLKs are characterized by a highly conserved N-terminal kinase domain (KD) and a noncatalytic C-terminal domain known as the PBD, which directs their subcellular locations and regulates the interaction of proteins. A total of five mammalian PLK family members have been identified: PLK1, PLK2/SNK, PLK3/FNK/PRK/CNK, PLK4/SAK/SZK18, and PLK5 (Raab et al. 2021). Among them, PLK1 is the most investigated member and has been found to be highly expressed in a variety of cancers (Lee et al. 2015). PLK1 is traditionally considered to participate in various mitotic events. Numerous studies have shown that PLK1 can drive the G2/M phase transition by controlling the activity of the CDK1/Cyclin B1 complex (Gavet and Pines 2010;Seki et al. 2008). In addition, PLK1 regulates cytoplasmic separation, mitotic spindle formation, and membrane formation in mitotic telophase by phosphorylating mitotic kinesin-like protein l (MKLP-1) (Liu et al. 2004). Based on a large number of studies, the mechanism of PLK1 is becoming increasingly clear during cell cycle progression. Recently, increasing evidences have shown that PLK1 also controls many nonmitotic events. For instance, X. Ma et al. found that PLK1 phosphorylates glucose-6-phosphate dehydrogenase (G6PD), thereby activating the enzyme and further upregulating the pentose phosphate pathway (PPP) by facilitating the formation of its active dimer . Inhibition of PLK1 has also been proven to be effective in limiting the progression of various cancers (Jeong et al. 2018;Dufies et al. 2021;Chen et al. 2019). However, as an attractive target of cancer therapeutics, little is known about the nonmitotic functions of PLK1 in CRC. Here, we focused on the effect of PLK1 in inducing endogenous mitochondrial apoptosis through the regulation of mitochondrial function in CRC. Apoptosis is considered a typical kind of programmed cell death and refers to the spontaneous and orderly death of cells to maintain homeostasis of the internal environment (Pistritto et al. 2016). Apoptosis dysfunction is responsible not only for tumor progression but also for tumor resistance to therapies (Wong 2011). Cell apoptosis is mainly mediated by the exogenous death receptor pathway and endogenous mitochondrial pathway (Fu et al. 2020). Specifically, members of the Bcl2 family, including proapoptotic and anti-apoptotic proteins, are involved in regulating the release of mitochondrial-related apoptotic factors in the mitochondrial pathway (Circu and Aw 2010;Shieh et al. 2010). Bax, a proapoptotic factor, increases the permeability of the mitochondrial membrane by translocating from the cytoplasm to the membrane (Jia et al. 2015). Once Bax translocates to the mitochondrial surface, apoptotic signaling is activated and initiates Cyt-c and apoptosis inducing factor (AIF) release from the mitochondria to the cytosol. Subsequently, Cyt-c binds to Apaf-1 and activates caspase-9. At the same time, activated caspase-9 cleaves the effector caspase-3, which triggers the cleavage of numerous key cellular substrates, such as PARP, and further induces apoptosis (Cory and Adams 2002;Li et al. 1997;Huang et al. 2017). In contrast, the anti-apoptotic protein Bcl2 prevents the caspase cascade, which counteracts Bax and binds to the outer mitochondrial membrane, thereby maintaining mitochondrial membrane integrity and enhancing cell survival (Brenner and Mak 2009;Tsujimoto and Shimizu 2007).
In this study, we confirmed the abnormal expression of PLK1 in CRC through database and IHC analyses of CRC patient tissues. These results indicated that PLK1 1 3 upregulation was involved in cancer progression. This is in line with the results of previous studies (Ran et al. 2019). Next, we found that inhibiting PLK1 by RNAi or the small molecule inhibitor BI6727 resulted in reduced cell viability, decreased cell migration and proliferation abilities, and obviously increased the number of apoptotic cells. Next, we further elucidated the underlying mechanism of CRC cell apoptosis induced by PLK1 inhibition. We found that PLK1 inhibition increased the accumulation of intracellular ROS, which might directly lead to the apoptosis of CRC cells. With reference to the results of previous studies, the increased ROS levels may be related to the mitochondrial apoptosis pathway (Liu et al. 2017b;Zhao et al. 2019). The application of the antioxidant DMTU demonstrated that increased ROS levels might lead to mitochondrial dysfunction, including a decreased ratio of Bcl2/Bax, loss of MMP, release of Cyt-c from mitochondria to the cytoplasm, caspase-3 activation, and finally induction of apoptosis of CRC cells. In addition, HCT116 xenograft tumor model was established to evaluate the effects of PLK1 in vivo. As expected, BI6727 significantly inhibited tumor growth compared with the control group. Remarkably, decreased tumor size was strongly related to the induction of cell apoptosis by BI6727 treatment. Collectively, these results demonstrated that inhibiting or silencing PLK1 could elevate intracellular oxidative stress, which resulted in the imbalance of Bcl2 and Bax expression and further destroyed mitochondrial function, thereby activating the mitochondrial apoptosis pathway in CRC. However, how PLK1 mediates the generation of ROS remains unknown. In turn, ROS accumulation may also be the result of mitochondrial dysfunction. Moreover, it is worth noting that Bcl2 protein also plays critical roles in nonapoptotic cellular processes, such as calcium homeostasis (Shalaby et al. 2020) and autophagy (Raab et al. 2021). Whether the decreased Bcl2 expression caused by PLK1 inhibition is associated with other biological functions needs to be further studied in CRC.
In conclusion, these findings indicate that PLK1 promotes the tumorigenesis of CRC by inhibiting ROS-mediated mitochondrial dysfunction. Moreover, the underlying mechanism of PLK1 inhibition-induced apoptosis further enriches our understanding of PLK1 function in CRC development. These findings also provide more evidences to support that BI6727 may be a novel potential therapeutic strategy in the treatment of CRC. As discussed above, the clinical application of novel approaches to directly target the mitochondrial apoptotic pathway may be highly effective in CRC treatment.