The human liver cell lines, LO2 and BEL7402, were purchased from the Cell Bank of Chinese Academy of Sciences. The HEK 293T cell line was gifted from Dr. Zhiyin Song (the College of Life Sciences, Wuhan University). Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Cat. SH30243.01, Hyclone, USA) supplemented with 10% fetal bovine serum (FBS, Cat. A0500-3011, Cegrogen, Germany) and 1% penicillin-streptomycin solution (Hyclone). Cells were cultured in a humidified incubator with 5% CO₂ at 37°C. For cell treatment, various doses of BI6727 (Cat. T6019, Topscience, Shanghai, China) were added into the cell culture medium, and cells were collected at the indicated time points to measure the growth rate by cell proliferation assays (see below).

For cell RNA extraction, cells were trypsinized by 0.05% trypsin-EDTA (Cat. S320JV, Shanghai BasalMedia Technologies Co., China) and collected by centrifugation with 500g for 3 min. For tissue RNA extraction, ~20 mg of tissue was homogenized and collected with centrifugation. Total RNA was isolated with TRIzol (Cat. 15596018, Thermo Fisher, USA). An aliquot of 1 μg of mRNAs was used for reverse transcription by HiScript II Q RT SuperMix (Cat. R223-01, Vazyme, Nanjing, China), and real-time quantitative PCR (RT-qPCR) was performed according to the manufacturer’s protocol (Cat. MQ10101S, Wuhan Monad Medicine Tech Co., China). The primer sequences used in quantitative PCR experiments are listed in Supplementary Table S1 .

WT or shSETD3 HEL7402 cells were collected, and total RNAs were extracted using TRIzol (Cat. 15596026, Invitrogen, USA) and further purified via chloroform extraction methods. RNA-Sequencing (RNA-seq) libraries were prepared using KAPA Stranded RNA-Seq Library Preparation Kit according to the manual (Cat. KK8401, KAPA Biosystems, USA), sequenced by Berry Genomics Company (Beijing, China) on Illumina NovaSeq 6000.

FastQC was used for quality control on raw sequence data. The filtered reads were normalized, and differential gene expression analysis was performed using DESeq2 v1.24.0. Gene Ontology was performed by DAVID v6.8. For heatmap, we used R package edgeR to calculate counts per million (CPM) of each gene; then, log (CPM) value was used to make heatmap by using R package ComplexHeatmap. R version 3.6.0 was used for some custom analysis.

Western blotting was performed as described previously. Briefly, cells were collected and lysed by radioimmunoprecipitation assay (RIPA) buffer [50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS)] with a protease inhibitor cocktail (Cat. B14001, Bimake, China). Tissues were grounded by an electronic homogenizer and lysed by RIPA buffer as well. Protein extracts were separated by SDS polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes (Cat. ISEQ00010, Millipore, USA). After blocking with 5% non-fat milk in TBST buffer [10 mM Tris–HCl, 150 mM NaCl, 0.5% (v/v) Tween-20, pH 7.5]. Membranes were probed with the following primary antibodies including mouse anti-SETD3 (1G7, mouse monoclonal, generated by Wuhan Dia-An Company), rabbit anti-PLK1 (Cell Signaling Tech, USA, 4535S), rabbit anti-pT210-PLK1 (Cell Signaling Tech., USA, 9062S), mouse anti-GAPDH (Proteintech, China, 60004-1-Ig), rabbit anti-β-actin (ABclonal, China, AC026), and rabbit anti-Flag (MBL, Japan, PM020). Membranes were washed and incubated with the appropriate secondary antibodies (Jackson ImmunoResearch Laboratory, USA) at room temperature for 1–2 h. Blotting signals were detected by ECL detection reagents (Cat. 36208ES76, Shanghai Yeasen Biotech, China). Prior results were captured by exposure with X-ray films (Fuji Company), and later results were captured by a chemiluminescence imaging system (ChemiScope 3000mini, Clinx Science Instruments Co. Ltd., Shanghai, China). For animal tissues, more than three individual mice samples per group were used for blotting. Densitometric quantification of protein bands was performed using ImageJ (NIH, Bethesda, USA).

Short hairpin RNA fragments (shRNAs) of mouse SETD3 containing 5′-GCTGGAGATCAGATTT ACATT-3′ (shSETD3) or mouse Plk1 containing 5′-GCACCGCAA TCAGGTCATTCA-3′ (shPlk1) were cloned into pLKO.1 vector using the restriction enzymes EcoRI and AgeI (New England Biolabs). Two shRNAs of human SETD3 containing 5′-GCTTTGGTTTGAGAGCAACAA-3′ or 5′-GAAGAAGATGAAGTTCGGTAT-3′ were cloned into plko.1 vector as well. Wild-type (WT) or catalytic inactive Y313A mutant of human SETD3 and WT or catalytic inactive K82M mutant of PLK1 were cloned into pCS2-3xFlag vector. For expression, plasmids were transfected into mammalian cells using lipofectamine 2000 (Cat. 11668019, Invitrogen, USA), according to the manufacture’s protocol. For generation of knockdown cell line, shRNA constructs were transfected into HEK 293T cells along with the helper plasmids pMD2G and psPAX2 to produce virus particles. Knockdown cells were selected by using puromycin and validated by Western blotting.

Cells were counted by Automatic Cell Counter (Guangzhou Bodboge Technology Co., Ltd.), and equal numbers of cells were seeded into six-well plates for 24 h, and cell layers were scratched using a 200-μl pipette tip to form wound gaps. Plates were washed twice with pre-warmed phosphate-buffered saline (PBS), and the cells were maintained in DMEM without FBS. The cells were photographed at 0 and 48 h to record the wound width. Three independent replicates were conducted and quantified.

The migratory capability was detected using transwell chambers (8 μm pores, BD Bioscience, Cat. 353097, USA). Briefly, 5,000 cells were plated on the surface of each upper chamber. The media containing 10% FBS was injected into the lower chambers to stimulate cell migration. After 36-h incubation, cells on the lower filter surface were fixed with 4% paraformaldehyde and then stained with 0.1% crystal violet stain solution (Cat. G1063, Solarbio, China). Images were taken by an optical microscope.

Cell proliferation assays were performed as described previously (13). Briefly, LO2 or BEL7402 cells with expressing SETD3 or PLK1 constructs or stably knockdown of SETD3 were treated with PLK1 inhibitor BI6727 at the indicated concentrations and measured using a Cell Counting Kit-8 (CCK-8, Cat. EC008, Engreen Biosystem Ltd., New Zealand) according to manufacturer’s instructions. Cell numbers in each well were normalized using Automatic Cell Counter (JSY-SC-031N, Guangzhou Bodboge Technology Co., Ltd.). Each time point was repeated in three wells, and the experiments were independently performed for three times. The effects of cell proliferation were characterized by a growth curve, which was plotted by the value of OD₄₅₀.

The 18 pairs of liver cancer tissues were collected from patients in Hubei Cancer Hospital. The liver cancer tissue microarrays (HLivH020PG02-M-209, Xi’an Alenabio Company, China), which contained samples from 30 pairs of human hepatocellular carcinoma and adjacent tissues, were purchased from Shanghai Outdo Biotech. Informed consents were obtained from all subjects in accordance with the protocol approved by the individual institutional ethics committees.

WT C57BL/6 mice (8 weeks old) were subjected to hydrodynamic injection procedures as described previously (25). Briefly, 1 μg pCMV-CAT-T7-SB100X, 6.4 μg pT3-Myr-AKT-HA, 6.4 μg pT3/Caggs-NRAS-V12, and either 6.4 μg pT3-U6-shCtrl or equal amount of pT3-U6-shSETD3 were mixed well and diluted in 2 ml of 0.9% NaCl, filtered and injected into the lateral tail vein of mice within 9 s. Five control mice were injected with 0.9% NaCl solution. After 4 weeks post-injection, all mice were sacrificed in groups to monitor liver tumorigenesis. Livers were separated from each mouse, and liver lesions were diagnosed macroscopically and several biochemical indexes described below.

Human tissue microarray slides were deparaffinized as described previously (13). For mice tissue staining, mice livers were isolated from the indicated male mice, rinsed in ice-cold PBS, fixed in 4% polyformaldehyde for 48 h at 4°C, and then dehydrated, embedded in paraffin, and sectioned. Tissue sections (4 μm) were used for staining. Slides were then incubated with the α-SETD3 or α-PLK1 antibody at 1:1,000, respectively, followed by biotin-conjugated goat anti-rabbit IgG antibody or goat anti-mouse IgG antibody at 1:10,000, respectively. Detections were performed using a 3,3′-diaminobenzidine (DAB) reagent (Cat. E-BC-K227-S, Elabscience, Wuhan, China), and nuclei were stained with hematoxylin before examination by a microscope. At least four mice per group were used for assay. The intensity score was determined by evaluating staining intensity of positive staining as described previously (13).

For serum triglyceride (TG) and total cholesterol (TC) content measurements, blood plasma samples were obtained just before necropsy and deposited for 30 min at room temperature. Serum was harvested by centrifugation at 3,000 rpm for 10 min at 4°C. Serum TG and TC contents were determined using the Triglyceride Assay Kit (Cat. A1101-1-1, NJJCBIO, Nanjing, China) and Total Cholesterol Kit (Cat. A1111-1-1, NJJCBIO, Nanjing, China), respectively. The operation was performed according to the manufacturer’s manual.

For measurement of the content of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), serum was harvested as described above. The ALT and AST contents were determined using the Alanine Aminotransferase Assay Kit (Cat. C009-2, NJJCBI, Nanjing, China) and Aspartate Aminotransferase Assay Kit (Cat. C010-2, NJJCBIO, Nanjing, China), respectively. The operation was performed according to the manufacturer’s manual.

The luciferase (LUC) reporter plasmid pGL4 vector was generously gifted from Dr. Xiao-Dong Zhang’s laboratory (the College of Life Sciences, Wuhan University). The ~2-kb fragment of PLK1 promoter extending −1,867 to +27 was amplified by PCR from genomic DNA of HeLa S3 cells as described previously (26) and cloned into pGL4 vector’s multiple cloning site just before LUC gene. For Luciferase reporter assays, 60 ng pGL4-PLK1 (−1,867/+27) reporter vector, 4 ng pRL-TK (Renilla), and 300 ng pCS2-3xHA-SETD3 were transfected into 293T cells. After 24–36 h, cells were lysed, and the luciferase activity was tested according to the manufacturer’s instructions (Cat. E1910, Promega, USA). The primer sequences used to generate various pGL4-PLK1 reporter constructs are listed in Supplementary Table S1 .

Chromatin immunoprecipitation (ChIP) assays were modified as described previously (27). Briefly, ~ 1 × 10⁷ cells were fixed with 1% formaldehyde and then quenched with glycine to a final concentration of 0.125 M. The cells were washed three times with PBS and then harvested in cell lysis buffer (5 mM PIPES, pH 8.0, 85 mM KCl, 0.5% NP-40) with the protease inhibitors and phenylmethylsulfonyl fluoride (PMSF) (1 mM final concentration). DNA in nuclei lysis buffer (50 mM Tris–HCl, pH 8.0, 10 mM EDTA, 1% SDS, and protease inhibitors) was sonicated for 20–25 cycles (high output, 30 s on, 30 s off) using the Bioruptor (Diagenode). After centrifugation, four volumes of ChIP dilution buffer (16.7 mM Tris–HCl pH 8.1, 167 mM NaCl, 0.01% SDS, 1.1% Triton X-100, and 1.2 mM EDTA) were added to the supernatant. The lysate was then incubated with anti-Flag affinity beads (SA042001, SMART LifeSci, China) at 4°C overnight. The beads were washed five times, and DNA was eluted by ChIP elution buffer (0.1 M NaHCO₃, 1% SDS, and 30 μg/ml proteinase K). The eluent was incubated at 65°C overnight, and DNA was extracted with DNA purification kit (Cat. DP214, Tiangen Biotech, Beijing, China). Purified DNA was analyzed by quantitative PCR.

Unless otherwise indicated, all data are presented as the mean ± SD from at least three biological replicates, and statistical differences between any two groups were compared by unpaired t-tests using Prism 7 software. The relative immunohistochemical staining (IHC) staining scores of human liver tissue microarrays were compared by unpaired Student’s t-tests. A p-value of <0.05, 0.01, 0.001, or 1 × 10⁻⁴ was considered statistically significant and marked as “*,” “**,” “***,” and “****,” respectively. “NS” indicates “not significant.”

All procedures were conducted in accordance with the Animal Guidelines for the Care and Use of Laboratory of Wuhan University (No. WDSKY0202001). All mice were housed, fed, and treated in accordance with protocols approved by the Animal Experimentations Ethics Committee of the Wuhan University.

Our previous study indicated that SETD3 level is upregulated in liver tumor BEL7402 cells, and overexpression of SETD3 led to mice xenograft tumorigenesis. To investigate underlying mechanism of how SETD3 involves in hepatocellular carcinogenesis in situ, RNA-Seq analysis were performed, and the differentially expressed genes (DEGs) in knockdown of SETD3 cells relative to WT BEL7402 hepatocellular tumor cells were analyzed ( Figure 1A ). We totally identified 1,344 DEGs (threshold set as 1.4-fold change, p < 0.05), including 590 upregulated genes and 754 downregulated genes overlapped at two pairs of shSETD3 cells vs. shControl cells ( Figure 1B ). Interestingly, gene ontology (GO) analysis indicated that the downregulated DEGs are mainly enriched in cell-cycle-related process, including cell division, cell proliferation, and cell cycle ( Figure 1C ). The gene set enrichment analysis (GSEA) showed that knockdown of SETD3 is closely correlated with lower expression of genes participated in cell cycle-related pathways ( Figure 1D ). Based on these data, we hypothesized that SETD3 might promote liver tumorigenesis mainly through regulating cell-cycle-related gene expression.

To further verify genome-wide RNA-seq data, expression levels of several top candidates were examined using RT-qPCR analysis. Consistent with whole-genome RNA-seq data, knockdown of SETD3 significantly reduced mRNA levels of several genes, including PLK1, BOP1, FOXC1, and PCNA. In contrast, mRNA level of SF3A2, a control gene, was not affected ( Figure 1E ). Consistently, we also observed that knockdown of SETD3 decreased PLK1 protein levels ( Figure 1F ). Meanwhile, we also chose liver immortalized LO2 cells with overexpression of SETD3 to phenocopy tumorigenesis of normal liver cell line. Consistently, we found that overexpression of SETD3 in LO2 cells increased mRNA levels of these genes except the control gene SF3A2 ( Figure 1G ). Moreover, overexpression of WT but not Y313A mutant of SETD3 enhanced PLK1 protein levels ( Figure 1H ). We also found that transient expression of WT SETD3, but not Y313A mutant, in shSETD3 cells was also capable of rescuing protein levels of PLK1 ( Figure 1I ). Altogether, these results indicate that SETD3 regulates PLK1 level in human hepatocellular tumor cells, likely in a catalytic-dependent manner.

A growing body of evidence demonstrated that hyperactive PLK1 is required for cell proliferation (28, 29). To investigate whether the effect of SETD3 on liver tumor cell proliferation relies on PLK1 kinase activity, cell growth assays were performed using SETD3-overexpressed LO2 cells treated with a PLK1 inhibitor BI6727. We observed that overexpressing WT, but not the Y313A mutant of SETD3, facilitates cellular proliferation compared to the control cells ( Figure 2A, a ). As expected, cells treated with higher doses of BI6727 (>20 nM) significantly attenuated cell proliferation, even supplemented with overexpression of WT SETD3 ( Figure 2A, b–d ). Western blot analysis showed an equal expression of SETD3 and PLK1 in the indicated cells ( Figure 2B ). Furthermore, the effect of PLK1 kinase activity on BEL7402 tumor cell proliferation was also examined. In agreement with previous results, knockdown of SETD3 retarded cell growth relative to the control cells ( Figure 2C, a ), whereas overexpression of WT, but not catalytic inactive K82M mutant of PLK1, promoted cell growth ( Figure 2C, b ). The difference of cell growth rates between WT and K82M mutant does not result from the divergent expression levels, as comparable PLK1 protein levels were displayed ( Figure 2D ). Interestingly, overexpression of WT PLK1, but not K82M mutant, can only partially restore decreased cell proliferation caused by knockdown of SETD3 in BEL7402 cells ( Figure 2C, c,d ), suggesting that PLK1 may only act as one of the downstream targets regulated by SETD3.

To explore whether SETD3 is important for cellular migration, wound healing and transwell assays were performed in both LO2 and BEL7402 cells. We noticed that overexpression of WT SETD3, but not Y313A mutant, promoted cell migration in normal liver LO2 cells ( Figures 3A, B ). In contrast, knockdown of SETD3 reduced cell migration in tumor liver BEL7402 cells ( Figures 3C, D ). Consistently, overexpression of WT but not Y313A mutant of SETD3 facilitated cell migration ability ( Figures 3E, F ), whereas knockdown of SETD3 reduced cell migration ( Figures 3G, H ). These data suggest the important role of SETD3 in liver cell migration.

Next, we want to know whether the effect of SETD3 on cell migration is dependent on PLK1 kinase activity. Wound healing assays showed that overexpression of WT PLK1, but not the K82M catalytic-dead mutant, promoted cell migration ( Figure 3I , top panel). However, knockdown of SETD3 was able to partially compromise phenotypes of accelerated cell migration caused by overexpression of PLK1 ( Figures 3I, J ). Meanwhile, cell migration assays also showed similar results, in which overexpression of WT PLK1, but not K82M mutant, increased cell migration, whereas knockdown of SETD3 can rescue the accelerated phenotype caused by PLK1 overexpression ( Figures 3K, L ). Together, we conclude that SETD3 is required for PLK1-promoted liver cell migration.

Next, we sought to investigate whether upregulated SETD3 and higher PLK1 levels are concurrent in vivo. Therefore, 18 pairs of human tumor and adjacent liver tissues were collected, and SETD3 and PLK1 levels were examined by Western blot analysis ( Figure 4A ). We observed that 15 out of 18 pair tissues displayed simultaneously increased SETD3 and PLK1 levels in tumor samples compared to their corresponding adjacent samples ( Figures 4B, C ). Correlation analysis indicated that upregulated SETD3 was positively correlated with higher levels of PLK1 in hepatocellular carcinoma samples (R = 0.4847, p = 0.021, Figure 4D ). To further consolidate this result, commercial purchased liver tumor tissue microarrays containing 24 pairs of human hepatocellular carcinoma and adjacent pairs of tissues were immunostained with α-SETD3 or α-PLK1 antibodies, respectively ( Figure 4E ). Consistently, we observed that higher levels of SETD3 are positively correlated with higher levels of PLK1 (R = 0.6623, p = 0.0006, Figures 4F–H ). Overall, we unveil a positive correlation of SETD3 and PLK1 in human liver tumors.

Several mouse experimental models have been widely used to simulate hepatocarcinogenesis, including xenograft model, diethylnitrosamine (DEN)-induced HCC model, oncogene-induced HCC model, and transgenic HCC model (30). Compared to DEN-induced or transgenic HCC model that needs over 6 months to induce HCC, an elegant and simple method was developed for liver-specific transgenesis that combines hepatic overexpression of the human oncogene AKT and NRAS and somatic integration mediated by Sleeping Beauty (SB) transposase system, which allowed generation of HCC transgenic model with reduced time and resource (25).

Thus, transposons encoding HA-tagged myristylated AKT (myr-AKT) plus NRAS-V12 (G12V) were mixed with plasmids expressing the SB transposase (named as SB-MIX1) in the presence of shControl or shSetd3 construct and then hydrodynamically delivered to the livers of 8-week-old C57BL/6 male mice via tail vein injection. A group of mice was injected with 0.9% NaCl solution as a control. Livers were harvested at 12 weeks post-hydrodynamic injection ( Figure 5A ). As expected, livers in the control mice group looks normal, whereas livers in the mice with myr-AKT plus NRAS transfected group showed a spotty and paler appearance and were substantially enlarged, suggesting a prominent tumor development ( Figures 5B, C ). However, livers in the mice group injected with myr-AKT plus NRAS plasmids with shSetd3 displayed a rather normal morphology and smaller liver sizes and showed compromised tumor development, although liver sizes were still 1.1-fold larger than the control group, but 3-fold smaller than the shCtrl group ( Figures 5B–D ). The protein levels of mouse Setd3 in different groups of liver tissues were examined by Western blot, and decreased Setd3 levels were only found in the liver tissue with transfection of shSetd3, but not found in the lung or heart tissue ( Supplementary Figures S1A, B ). Immunochemical staining confirmed that Setd3 levels did not alter in mice transfected with myr-AKT plus NRAS plasmids and were visually decreased in mice cotransfected with shSetd3 and myr-AKT plus NRAS constructs ( Supplementary Figures S1C, D ). Consistently, the total cholesterol (TC) and total triglyceride (TG) levels in serum were significantly increased in the myr-AKT plus NRAS group compared to the control group, but decreased upon transfection of shSetd3 construct ( Figure 5E ). Moreover, markers of liver injury, such as ALT and AST in serum, also remarkably increased in the myr-AKT plus NRAS group, but restored to the control levels when co-transfected shSetd3 construct ( Figure 5F ). These results indicate that Setd3 is critical for hepatocellular tumor development.

To examine whether Plk1 indeed acts at the downstream of Setd3-induced hepatocarcinogenesis in vivo, the same hydrodynamics-based injection procedure was utilized ( Figure 6A ). In agreement with the observations shown above, overexpression of Setd3 in the presence of myr-AKT plus NRAS constructs facilitated liver tumor development, accompanied with obvious spotty and paler appearance and larger liver sizes, compared to the control group ( Figures 6B–D ). As expected, knockdown of Plk1 co-transfected with Setd3 and myr-AKT plus NRAS constructs was able to restore normal liver morphology and liver sizes ( Figures 6B–D ). The Plk1 knockdown efficiency and overexpressed Setd3 levels in different mice groups were validated by Western blot analysis and IHC staining ( Supplementary Figure S2 ). We also examined several body indexes that present the liver status in these mice. In line with previous results, serum TC and TG levels and AST and ALT levels were significantly increased in mice with overexpression of Setd3 relative to the control group, but were restored in mice with shPlk1 construct ( Figures 6E, F ). Taken together, we speculate that Plk1 very likely functions at the direct downstream of Setd3 in hepatocarcinogenesis.

To testify our assumption, we examined whether SETD3 directly regulates PLK1 transcriptional level using dual-luciferase reporter assays. We engineered the PLK1 promoter containing ~2,000 bp upstream of PLK1 coding region into 5′ end of LUC gene encoding luciferase as described previously (26) and tested the effect of SETD3 on the activation of LUC gene expression. We noticed that the expression of WT SETD3 was able to activate luciferase activity, and the catalytic activity of SETD3 is required for this activation, which does not result from lower protein level of SETD3 Y313A mutant compared with the WT ( Figures 7A, B ). To narrow down the targeted region on PLK1 promoter by SETD3, we systematically truncated the PLK1 promoter and generated a series of LUC reporter constructs ( Figure 7C ). These reporter constructs were stably expressed in HEK 293T cells, and a 3xHA-SETD3 plasmid was individually transfected into the indicated cells. The relative luciferase activity was examined after 24 h transfection. We noticed that once there is removal of the region of −1,376/−1,139 bp on the PLK1 promoter, the activation of luciferase expression was impaired ( Supplementary Figure S3 ). Intriguingly, if only retaining the nearest 242 bp of the PLK1 promoter, the luciferase activity was even higher than the longest LUC reporter construct ( Supplementary Figure S3 ). To rule out a possible heterogeneous effect using different types of cell line, the dual luciferase reporters were co-transfected with 3xHA-SETD3 into immortalized liver LO2 cells. Consistent with previous results, the removal of the region of −1,376/−1,139 bp on the PLK1 promoter compromised the luciferase activity, whereas the nearest 242 bp of the PLK1 promoter was also required to activate luciferase gene expression ( Figure 7D ). We confirmed that the epitopic-expressed SETD3 protein levels were relatively comparable in each stable cell line ( Figure 7E ). To validate which region of the PLK1 promoter is more critical for SETD3 to activate luciferase activity, we constructed several truncated forms of LUC reporter gene that lack the indicated DNA fragments ( Figure 7F ), and relative luciferase activity was examined. As shown in Figure 7G , deletion of the second region (Δ2, −1,376/−1,139 bp), but not the other two regions (Δ1, −1,856/−1,630 bp; Δ3, −835/−610 bp), is critical for its activation by SETD3. We also confirmed that SETD3 protein levels were comparable in these samples ( Figure 7H ). Intriguingly, deletion of the nearest promoter region of PLK1 (Δ4, −201/+0 bp) abolished a basal luciferase activity compared to that expressing the empty pGL4 vector alone, suggesting that this region may be essential for binding of certain transcriptional factor. To confirm that SETD3 regulates PLK1 expression by association of the PLK1 promoter, ChIP-qPCR analysis was performed. We observed that, compared to the enrichment of SETD3 at the VEGF promoter that served as a positive control reported previously, the enrichment of SETD3 at the promoter of PLK1 gene was also obvious, especially the second but not the fourth region of the promoter, indicating a potential role of SETD3 in PLK1 expression ( Figure 7I ). Altogether, we conclude that SETD3 could target PLK1 locus at the −1,400/−1,000 bp region and facilitate its expression.