STAT3 regulates miR93-mediated apoptosis through inhibiting DAPK1 in renal cell carcinoma

Signal transducer and activator of transcription 3 (STAT3) is an essential member of the STAT family. STAT3 regulates diverse genes that mediate inflammatory reactions, cell survival, proliferation, and angiogenesis, and it is aberrantly upregulated and activated in various types of malignancies. Furthermore, STAT3 signalling is involved in multiple feedback loops and pathways. In this study, we demonstrate that miR-93-3p plays an oncogenic role in renal cell carcinoma (RCC) by enhancing RCC cell proliferation and suppressing apoptosis. In addition, STAT3 can regulate the transcription of miR-93 by directly binding its promoter region. miR-93 can inhibit death-associated protein kinase 1 (DAPK1) at the protein level. Moreover, STAT3 can block DAPK1 expression at the RNA level. Importantly, we verified that DAPK1 overexpression in turn suppresses the entry of activated STAT3 into the cell nucleus. Thus, this study reveals a potential continuously activated signalling transduction pathway, STAT3-miR93-DAPK1, and may provide a novel clinical therapeutic approach for RCC.


Introduction
Renal cell carcinoma (RCC) accounts for 2-3% of all adult malignancies in China; in term of incidence, RCC ranks third among the urinary tumours, following prostate and bladder carcinoma. Furthermore, the incidence of RCC has increased by 6.5% per year over the past 20 years, with 40% of RCC patients dying. Early diagnosis confers the best chance for curing RCC. However, patients with early-stage RCC lack typical clinical symptoms, such as pain, the presence of a mass, or haematuria. Unfortunately, more than 30% of RCC patients have metastatic lesions once diagnosed [1][2][3]. Therefore, the development of efficient clinical diagnostic strategies is critical for the prevention and management of RCC.
MicroRNAs (miRNAs) are a class of highly conserved endogenous small non-coding RNAs present in almost all organisms. They function as negative regulators of gene expression by base pairing and binding the 3′untranslated region (3′-UTR) of target mRNA, causing RNA degradation or translation suppression. Accumulating evidence has demonstrated that miRNAs play essential roles in biological processes, such as differentiation, proliferation, and apoptosis. The dysregulation of miRNAs is recognized as a crucial factor in the development and progression of diverse cancers [4][5][6][7][8].
The miR-106a-25 cluster, embedded within the 13th intron of the Mini-chromosome Maintenance protein 7 (MCM7) gene, encodes three miRNAs: miR-106b, miR-93, and miR-25 [9,10]. miR-93 is upregulated in many human cancers, such as gastric, breast, prostate, and ovarian tumours [11][12][13][14], and acts as an oncogene to modulate cell apoptosis, the cell cycle, and proliferation. However, the functions of miR-93-3p in RCC and its associated with signalling pathway remain unknown. The purpose of this study was to explore the function of miR-93 signalling pathway in cell apoptosis and proliferation and to elucidate its regulatory mechanisms and associated key molecules. Furthermore, we aimed to provide potential targets and biomarkers for RCC diagnosis and therapy.

Cell lines
Human renal carcinoma cell lines (786-O, ACHN, 769-P, OSRC-2, and CAKI-1 cells) and a proximal tubule epithelial cell line (HK-2 cells) were obtained from the Committee on Type Culture Collection Cell Bank of the Chinese Academy of Sciences (Shanghai, China). All the cell lines were authenticated and identified no mycoplasma contamination. All media were supplemented with 100 U/ml penicillin and 100 mg/ml streptomycin (TBD Science, Tianjin, China), and cells were cultured at 37°C in 5% CO 2 .

Tissue samples and clinical data collection
In this study, we analyzed 96 patients diagnosed pathologically with RCC at the First Hospital of Chinese Medical University from 2016 to 2017 who underwent surgical resection without chemotherapy or radiation. Adjacent normal kidney tissues samples were collected at a distance of more than 3 cm from the tumour and stored at −80°C until use. All tissues were processed for histological examination. The study was approved and supervised by the Ethics Committee on Human Research of the First Affiliated Hospital of Chinese Medical University, and all 96 patients who volunteered for this study provided their written informed consent.

Lentivirus vector transduction
Cells were transduced with STAT3-overexpressing lentiviruses (GenePharma Corporation, Suzhou, Jiangsu, China) in complete medium containing 5 μg/ml polybrene, incubated at 37°C in 5% CO 2 and cultured for 48 h. After cell harvesting, transduction efficiency was detected by qRT-PCR, and GFP expression was observed by fluorescence microscopy at 72 h after transduction. The cells were selected using fresh medium containing 2 μg/mL puromycin. STAT3-overexpressing lentiviruses were used in dual luciferase assay, ChIP assays, and immunofluorescence assay.

Xenograft tumour model and IHC staining
BALB/c nude mice (4 weeks old, 14-16 g, female) were purchased from Beijing Vital River Experimental Animal Technology Co, Ltd. and housed in a barrier facility on a 12hour light/dark cycle. The Institutional Animal Care and Use Committee of China Medical University supervised and approved all of the experimental procedures. According to the body weight, the mice were randomly divided into two groups. Each group contained five mice, which is experimentally estimated to be enough to detect the growing difference of denigrated tumours between two groups. miR-93-3poverexpressing ACHN cells and negative control cells (4-6 × 10 6 ) were inoculated subcutaneously in the left dorsal flanks of the mice. Tumour length and width were measured with a calliper, and tumour volume was calculated using the equation (length × width 2 )/2. The volumes of xenograft tumours were recorded every week. During the injection and measurement, the operators were blinded to the mice allocation. On the 45 th day, the animals were euthanized, and the tumours were excised, weighed, and paraffin-embedded. Serial 6.0 μm sections were cut and subjected to staining assays. Cells stained with Ki-67 (Abcam, Cambridge, MA, USA) were counted, and their proportion was determined to assess the proliferative capacity. Apoptosis was assayed by the DeadEnd

Cell proliferation and viability assay
Cells were seeded before treatment, and a Cell Counting Kit-8 assay (CCK-8, Dojindo Molecular Technologies, Inc. Shanghai, China) was used to assess the proliferation potential. Each assay was repeated at least three times. Three replicates were prepared for each time point.

Cell apoptosis assay
Cell apoptosis was examined using flow cytometry analysis with a FACSCalibur flow cytometer (Becton Dickinson Biosciences, San Jose, CA) equipped with CellQuest software (BD Biosciences). The cells were collected and stained with FITC Annexing V and PI (Pharmingen Annexin V FITC Apoptosis Kits, BD, San Diego, CA, USA).

IL-6 and stattic treatment
Interleukin 6 (IL-6, Sino Biological, Beijing, China) and Stattic (Selleck, Shanghai, China) were reconstituted according to the manufacturer's protocol. Before treatment, cells were cultured with serum-free medium for 12 h. The IL-6 and Stattic solutions were diluted in serum-free medium to different concentrations. Serum-free medium was used for the control group.

Immunofluorescence
Immunofluorescence images were captured with an inverted fluorescence microscope (Olympus, Tokyo, Japan). After transfection, cells were seeded in 24-well plates and treated with IL-6 at 20 ng/ml for 20 mins. Then, the cells were reacted with rabbit polyclonal antibody against pSTAT3 (1:200, #9145, Cell Signaling Technology, Danvers, MA, USA) in blocking buffer overnight. Before viewing, the nuclei were stained with DAPI (Beyotime, Shenzhen, Guangdong, China).

Dual-luciferase reporter assay
The GP-mirGLO plasmids in the first dual-luciferase reporter assay were constructed with the genomic region of human DAPK1, and they contained the wild miR-93-3p binding site and the mutant miR-93-3p binding site (Gen-ePharma Corporation, Suzhou, Jiangsu, China). The pGL3 plasmids in the other dual-luciferase reporter assay were constructed with the genomic region of human miR-93 promotor containing the wild STAT3 binding site and the mutant STAT3 binding site (GenePharma Corporation, Suzhou, Jiangsu, China), and the Renilla luciferase reporter plasmids (GenePharma Corporation, Suzhou, Jiangsu, China) were co-transfected and used as an internal reference in the second dual-luciferase reporter assays. Fortyeight hours after transfection, the cells were lysed in passive lysis buffer, and the Firefly and Renilla luciferase activities were tested using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA) according to the manufacturer's instructions. The firefly luciferase activity results were normalized to the Renilla activity results.

Chromatin immunoprecipitation assay
A chromatin immunoprecipitation (ChIP) assay was performed using the SimpleChIP® Plus Sonication Chromatin IP Kit (Cell Signaling Technology, Danvers, MA, USA) according to the manufacturer's protocol. The samples were added the immunoprecipitating antibodies (pSTAT3, 1:100, sc-8059, Santa Cruz, Dallas, Texas, USA; IgG, 1:200, #3900S, Cell Signaling Technology, Danvers, MA, USA) and incubated overnight at 4°C with rotation. After reverse cross-linking by heating at 60°C and vibration for more than 2 h, qRT-PCR was performed using the following promoter-specific forward and reverse primers: F: 5′-AAAACAAATTCCACGCTCCT-3′; R: 5′-GCTCTGCCA CTTCCTCACA-3′. The precipitated DNA was also amplified for 25 cycles and resolved on a 2% agarose gel to evaluate the target DNA.

Statistical analysis
Each experiment was repeated three times. All statistical analyses were performed using SPSS 21.0 statistical software (SPSS Inc., Chicago, IL, USA), and the results are presented as the mean ± SD. A two-tailed Student's ttest was used to evaluate the significance of differences between data from two groups for all pertinent experiments. The Mann-Whitney U test was performed for unpaired group comparisons. The correlation between T stage and miR-93 expression was calculated by Fisher's exact test while correlations between miR-93 expression and the other clinical characteristics were assessed by the chi-square test. All the data involved in t-test were normal distribution. The variation within each group was similar. A P-value < 0.05 was used to indicate significance. *P < 0.05.

miR-93-3p was upregulated in RCC tissues and cell lines
First, we assessed the expression levels of miR-93-3p in RCC tissues using online public data from The Cancer Genome Atlas (TCGA). We found that miR-93-3p expression levels were significantly upregulated in RCC tissues compared with normal tissues (Supplementary Fig. a). Furthermore, the expression levels of miR-93-3p in RCC samples and adjacent normal tissues were verified by qRT-PCR and normalized to U6 (P < 0.05; Fig. 1a). The results were consistent with the TCGA data analysis. Moreover, we assessed the miR-93-3p expression in RCC cell lines (ACHN, 786-O, 769-P, OSRC-2, CAKI-1), and a normal tubular epithelial cell line (HK-2). miR-93-3p expression was also found to be elevated in all of the RCC cell lines compared with the normal cell line (P < 0.05; Fig. 1b). As shown in Table 1, differences in clinical characteristics were analyzed between the high and low miR-93-3p expression groups, and they were obtained through dividing by the median relative of the miR-93-3p expression value. miR-93-3p expression was correlated with T stage and pathological grade, and less differentiated RCC tissues and tissues from RCC at an advanced T stage exhibited much higher miR-93-3p expression.

miR-93-3p increased RCC cell proliferation in vitro and enhanced tumour growth in vivo
To investigate the function of miR-93-3p in tumour development and progression, the RCC cell lines ACHN and 786-O were transfected with the agomir and antagomir of miR-93 ( Supplementary Fig. b). After transfection, we performed colony formation and CCK-8 assays to evaluate the effect of miR-93-3p on cell proliferation and viability. Overexpression of miR-93-3p increased the proliferation and viability of RCC cells while inhibition of miR-93-3p expression suppressed cell proliferation and viability (P < 0.05; Fig. 1c, d).
To examine the effects of miR-93-3p on the progression of RCC in vivo, nude mice were subcutaneously inoculated with ACHN cells. After injection of the cells transfected with the agomir of miR-93-3p or negative control, the two groups of mice were return to their cages and housed for 45 days. The tumours that received injection of the agomir grew much faster than the tumours from the negative control groups (Fig. 1e). The xenograft tumours from the agomir group were significantly larger than those from the control group. Further histological examination showed that the xenograft tumour cells in the miR-93 overexpression groups exhibited increased expression of Ki67 compared with that in the control groups (Fig. 1f). These results further confirmed that the upregulation of miR-93 enhanced tumour growth in vivo and in vitro.

miR-93 mediated RCC cell apoptosis by inhibiting DAPK1 expression
To evaluate the effect of miR-93 on the apoptosis of RCC in vitro, a flow cytometry analysis was performed. The proportion of apoptotic cells transfected with the miR-93 agomir was significantly decreased compared with that in the negative control group while inhibition of miR-93 expression promoted cell apoptosis (P < 0.05, Fig. 2a). Moreover, an in vivo TUNEL assay showed fewer apoptotic cells, visualized as green spots, in the agomir group than in the negative control group (Fig. 2b).
Next, we explored the potential targets of miR-93 with miRWalk 2.0 (http://diana.imis.athena-innovation.gr/Diana Tools/index.php?r=microT_CDS/index) and found that death-associated protein kinase 1 (DAPK1) is a putative target of miR-93-3p (Fig. 2c). Next, we investigated whether miR-93 can specifically regulate DAPK1 expression through qRT-PCR and western blotting. Western blotting confirmed that in the cells transfected with the agomir, upregulation of miR-93 significantly decreased the protein levels of DAPK1, while the two RCC cell lines transfected with the antagomir showed the opposite effect (P < 0.05, Fig. 2d). However, the qRT-PCR results showed no difference in DAPK1 level among the transfected groups (ns, P > 0.05, Supplementary Fig. c). This indicates that miR-93 might impact DAPK1 at the protein level but not at the mRNA level. To validate this result in vivo, we examined the DAPK1 expression in xenograft tissues using IHC and found that DAPK1 expression was downregulated by the agomir of miR-93 ( Supplementary Fig. d, e). Then, a dualluciferase reporter assay was performed to examine whether miR-93 directly binds the 3′-UTR region of DAPK1 mRNA. Based on the predicted binding site, we designed pmirGLO luciferase reporter plasmids containing wild-type (WT) or mutant-type (Mut) DAPK1 3′-UTR and cotransfected these plasmids with miR-93 agomir or negative control into 293T cells. The difference in absorbance ration showed downregulated luciferase activity in the cells transfected with WT plasmid and agomir. In contrast, no significant change in luciferase activity was found when the cells were transfected with Mut plasmids (P < 0.05, Fig. 2e). Then, we performed a recovery assay by western blotting to further verify that miR-93 regulates the protein expression of DAPK1 (Fig. 2f). We co-transfected DAPK1 and miR-93, and conducted CCK-8 and apoptosis recovery assay of the mock, DAPK1 and recovery groups to verify that the anti-apoptosis and pro-proliferation of miR-93 was achieved by binding and down regulating DAPK1 (Fig. 2g,  h). The results showed that the cells of the co-transfected group grew more vigorously, and the percentage of apoptotic cells was less than that cells of only the DAPK1transfected group.

STAT3 induced miR-93 expression
We further investigated whether miR-93 regulation is related to some crucial signalling pathways involved in cancer development or progression. First, we used the miR-93 gene to search the UCSC Genome Browser online database (http://genome.ucsc.edu/). We selected sequences upstream by 2000 bases and downstream by 1000 bases and obtained the promoter region sequence of miR-93. Then, we used the sequences as a search term for Jaspar (http://jaspar.genereg. net/) and Promoter Scan (http://www-bimas.cit.nih.gov/ molbio/proscan/) with the threshold score set at 90% and obtained a series of transcription factors and their binding sites. The bioinformatics data indicated that a STAT3binding site within the miR-93 promoter matched the results from the two websites (Fig. 3a). To verify this finding, we measured the expression of miR-93-3p and STAT3 in 96 RCC samples, which showed a positive correlation between the expression of these two factors (P < 0.05, R 2 = 0.41, Fig. 3b). Then, miR-93-3p expression in ACHN and 786-O cells under two adverse stimuli was quantified by qRT-PCR. We examined the effect IL-6 as an exogenous activator of STAT3 and Stattic as an exogenous inhibitor of STAT3 by western blotting. Then, we measured the concentration-and time-dependent effects of an activator and inhibitor on pSTAT3 formation by western blotting. The 786-O cell line was treated with an IL-6 solution at different concentrations (10 ng/ml, 20 ng/ml, 30 ng/ml and 40 ng/ml) for 20 mins. The levels of pSTAT3 and active STAT3 in the group treated with 20 ng/ml, 30 ng/ml and 40 ng/ml IL-6 were higher than those in the group treated with 10 ng/ml IL-6, and the differences in these values among the groups treated with the three higher IL-6 concentration were nonsignificant. After incubation with the 20 ng/ml IL-6 solution for 10 mins, the pSTAT3 level began to increase until 20 mins, when it peaked at twice the level in the group incubated in serum-free medium. Finally, we found that miR-93-3p expression was upregulated in 786-O cells treated with 20 ng/ml IL-6 and it increased 30 mins later and peaked after 2 h of treatment (Fig. 3d). Then, miRNA expression decreased gradually and returned to its normal level 6 h later. In the same way, we set different concentrations and time curves for Stattic, and found that miR-93-3p expression was downregulated under 20 µM Stattic stimulation; interestingly, it decreased 30 mins later, reached the lowest level after 2 h of treatment, and returned to its normal level 6 h later (Fig. 3d). Furthermore, we next directly knocked down STAT3 by transfecting two small interfering RNAs (siRNAs) to silence STAT3 and performed qRT-PCR to examine the change in miR-93-3p. Consistent with the effect of exogenous small-molecule compounds, miR-93 expression was found to be significantly decreased upon transfection with siRNA (Fig. 3e). These results demonstrated that the activation of STAT3 induced miR-93 expression, whereas the inhibition of STAT3 signalling suppressed miR-93 expression.

STAT3 directly bound the promoter region of the miR-93
To verify the direct binding of pSTAT3 at the predicted potential binding sites in the promoter region of miR-93, dual-luciferase reporter plasmids containing the pSTAT3 binding site sequence on the miR-93 promoter (MT) or corresponding mutant sequence (Mut), were constructed (Fig. 4a). After STAT3-overexpressed cells were transfected with the Mut and WT plasmids, they were cultured for 36 h and treated with 20 ng/ml IL-6 or serum-free medium for 2 h. Then, we washed the four groups of cells and performed a luciferase reporter assay. The dualluciferase reporter assay showed that IL-6 treatment significantly increased the luciferase activity in the cells transfected with the WT plasmids (Fig. 4b), demonstrating that pSTAT3 might bind with the predicted promotor region and play a translation role. Furthermore, the ChIP assay revealed that activated STAT3 bound with the promoter sequence of the miR-93, and these results were verified through qRT-PCR and DNA-agarose gel electrophoresis (Fig. 4c, d). The STAT3-overexpressed cells were used in ChIP assays. According to the predicted binding site, we designed one pair of primers that could be amplified for 196 bp DNA fragment containing the binding site. After the DNA fragments were purified by immunoprecipitation, Fig. 2 miR-93 suppressed the protein level of DAPK1 to downregulated RCC cells apoptosis. a The apoptosis cell count calculated by flow cytometer different groups. b The apoptosis cells marked with red arrows in the xenograft tumor shown by Tunel assays. c The highly conserved DAPK1 3′UTR and predicted miR-93-3p target sequence in the 3′-UTR of DAPK1 and the mutant type with 8 altered nucleotides. d Overexpression of miR-93-3p significantly decreased protein levels of DAPK1. e The luciferase assay was performed with co-transfection of miR-93-3p and wild-type or mutant-type DAPK1 3′-UTR. Firefly luciferase activity of each sample was normalized against renilla luciferase activity. f-h The recovery assays confirmed miR-93-3p promoted proliferation and inhibited apoptosis by mediating DAPK1. Fig. 4 STAT3, as a transcription factor, bound with the promoter region of miR-93. a miR-93 promoter region was cloned into a pmirGLO luciferase reporter plasmid. For the mutant type, 10 nucleotides at the predicted binding site were altered simultaneously. b Luciferase assay results showed that activation of STAT3 significantly increased the luciferase activity in cells transfected with luciferase reporter plasmid containing the wild-type miR-93 promoter. c qRT-PCR was also performed to measure the enrichment of predicted binding fragments. d ChIP assay results showed that STAT3 physically bound with the miR-93 promoter. Lane 1, input chromatin prior to IP. Lane 2, IP with the non-specific antibody, IgG. Lane 3, IP with pSTAT3 antibody when IL6-stimulating. Lane 4, IP with just pSTAT3. PCR was performed. In the IgG group, no peak was found in the PCR amplification curve, but it was obvious in STAT3 precipitation group, and the peak was the fastest in the IL-6 treatment group. These data strongly supported the notion that STAT3 directly induces miR-93 by enhancing its transcriptional expression.

DAPK1 and STAT3 interact with each other to regulate cell apoptosis and proliferation
In a previous study, knockdown of DAPK1 attenuated the curcumin-induced G2/M cell cycle arrest and apoptosis by modulating STAT3 in glioblastoma multiforme (GBM) cells [15]. We further investigated whether an unknown regulatory mechanism exists involving STAT3 and DAPK1 in RCC. We assessed DAPK1 expression in 786-O and ACHN cells transfected with STAT3 siRNA by qRT-PCR and western blotting and DAPK1 expression under Stattic stimulation (Fig. 5a, b, c). The results showed that DAPK1 was upregulated at both the protein and mRNA expression levels following exogenous cytokine stimulation and RNA interference cells. This finding demonstrated that STAT3, an oncogenic transcription factor, might regulate DAPK1 mRNA and protein expression in RCC. However, no difference in STAT3 was found at the protein or mRNA expression level in RCC cells transfected with pcDNA-DAPK1 ( Supplementary Fig. f, g). However, the level of pSTAT3 in the nuclei of cells transfected with the DAPK1overexpressed plasmids was significantly decreased compared with that in the negative control groups (Fig. 5d). Moreover, pSTAT3 entry into the nucleus was decreased in each image from the immunofluorescence assay in STAT3overexpressed RCC cells transfected with plasmids compared with RCC cells transfected with the negative control (Fig. 5e). Images of the DAPK1-overexpression group under a high-power field showed a ring of fluorescent signal around the nucleus that might be from accumulated pSTAT3 due to inhibited DAPK1 expression, but this signal was not observed in the group treated with IL-6. Thus, DAPK1 attenuated STAT3 transcriptional activity by decreasing pSTAT3 entry into the nucleus. Alternatively, inactivation or activation of STAT3 led to an increase or decrease, respectively, in DAPK1 at the mRNA and protein levels.

Discussion
Early diagnosis plays a crucial role in RCC treatment. Hence, highly specific and sensitive diagnostic biomarkers for RCC are necessary and urgently needed. In this study, we identified miR-93-3p as an onco-miRNA involved in the apoptosis and proliferation of RCC and revealed the potential STAT3/miR-93-3p/DAPK1 regulatory pathway. Our data demonstrated that activated STAT3 promotes miR-93-3p expression by binding its promoter region and that miR-93-3p suppressed the protein level of DAPK1 in RCC cells. Moreover, DAPK1 suppressed activation of the STAT3 pathway through blocking pSTAT3 translation and translocation into the nucleus. These steps generate a feedback regulation loop. Moreover, STAT3, as a transcriptional factor, induced the protein and mRNA expression of DAPK1, suggesting that STAT3 might directly regulate DAPK1 and dose not just impact DAPK1 mRNA through miR-93-3p (Fig. 6). DAPK1, a Ca 2+ /calmodulin (CaM)-dependent serine/ threonine-protein kinase, plays an important role in diverse apoptosis and autophagy pathways and immune responses in autoimmune disorders, neurodegenerative diseases, ischaemic damage and many types of cancer. It is necessary to reveal the complex regulation network of DAPK1 for studying tumorigenesis, inflammatory carcinogenesis and targeted drug development. Several mechanisms account for DAPK1 deregulation in cancer, including transcriptional and post-transcriptional regulation. The 5′-UTR of the DAPK1 gene contains CpG islands. Hyper-methylation of DAPK1 in CpG islands leading to gene silencing has been detected in many tumours [16]. DAPK1 expression was correlated with p53 activation in both normal and cancer cells in response to DNA damage. Electrophoretic mobility shift assays (EMSAs) and ChIP assays revealed that p53 can bind upstream of the first exon or within the first intron of DAPK1 and confirmed that p53 could positively regulate DAPK1 expression [17]. In additional to p53, ERK can control DAPK1 by phosphorylating DAPK1 at Ser735, promoting DAPK1 activity and triggering further DAPK-ERK interaction through their death domains, finally promoting cell apoptosis [18]. UNC5B interacts with DAPK1 through its death domains to exert its pro-apoptotic function. Moreover, UNC5B activates DAPK1 by inhibiting DAPK1 auto-phosphorylation at Ser308 [19]. At the mRNA level, DAPK1 downregulation can also be mediated by a miRNA-dependent mechanism. miR-103/107 can target the DAPK1 3′-UTR to interfere with translation and promote metastasis in colorectal cancer [20]. On the other hand, post-translational regulation of DAPK1 includes its protein phosphorylation by other kinases, auto-phosphorylation at Ser308, ubiquitination [21], and protease-mediated degradation.
STAT3 is an angiogenesis-related transcription factor related to RCC proliferation, migration, and survival [22][23][24]. However, it is unclear whether STAT3 interacts with DAPK1 in RCC. In this study, we regulated the expression or activation of STAT3 with siRNA and exogenous activator in RCC cell lines and clearly found that the mRNA and protein expression of DAPK1 was regulated, unlike that of STAT3. This finding demonstrated that STAT3 activation and upregulation might transcriptionally repress DAPK1. A previous report showed that the DAPK1 mRNA levels are negatively regulated via the non-canonical Flt3lTD/NF-κB pathway [25]. However, no reports related to STAT3 as a transcription factor that targets DAPK1 have been published. Hence, we searched bio-databases to determine whether STAT3 can bind the promoter region of DAPK1. A sequence analysis of the DAPK1 promoter (Database of Transcriptional StartSites (DBTSS): NM_004938) and the UCSC Genome Browser revealed that the DAPK1 promoter contains two putative STAT3-binding sites, region 1 (−1471 to −1821) and region 2 (−351 to −631). The preliminary findings showed that though its suppressive function, DAPK1 suppresses the TCR-and LPS-triggered NF-κB activation pathway [26]. The lungs and macrophages of DAPK1-knockdown mice secreted higher levels of IL-6 and CXCL1 in response to LPS than those of control mice [27]. Thus, DAPK1 can inhibit the activation of inflammatory cytokines and suppress the progression of the inflammatory reaction. Abnormal activation of inflammatory cytokines also plays important roles in tumorigenesis, progression, invasion, and metastasis. We hypothesized that DAPK1 affect inflammatory pathways and that this effect involves STAT3 or pSTAT3. Furthermore, we upregulated DAPK1 expression by plasmid transfection and found no changes in STAT expression at either the mRNA or protein level. Interestingly, the expression of miR-93-3p, which is downstream of STAT3, was increased. We investigated whether DAPK1 affects the activation of STAT3 or the transportation of pSTAT3 through an immunofluorescence analysis of the protein level of pSTAT3 in the nucleus. The progression of pSTAT3 transportation was suppressed by the increase in DAPK1 in RCC cells. Fig. 6 STAT3 regulates miR93-mediated apoptosis through inhibiting DAPK1 in Renal cell carcinoma. STAT3 promotes miR-93-3p expression by binding to promoter region, then miR-93-3p suppressed DAPK1. What's more, DAPK1 mediates the activation of STAT3 pathway through blocking pSTAT3 translation into the nucleus, and STAT3 decreased the expression of DAPK1.
In summary, activated STAT3 might be enriched in the DAPK1 promoter sequence and repress the DAPK1 transcription progress. On the other hand, DAPK1 might act as a negative regulator of STAT3 and attenuate the activity of STAT3 through preventing the pSTAT3 nuclear localization pathway. How does DAPK1 de-activate STAT3? We believe that there might be two possible mechanisms of this effect. The first is that DAPK1 impacts the protein conformation of activated STAT3 via protein-protein interactions. This leads to the dimerization of pSTAT3, which is the functional STAT3 structure, being blocked out of the nucleus. However, we performed Co-IP assays, which showed no direct binding of pSTAT3 and DAPK1 (Supplementary Fig. h). In the second mechanism, DAPK1 inhibits the upstream inflammatory factors, such as IL-6, TNF-α, and IFN-γ, which can activate STAT3 or translocate STAT3 to the nucleus. The most important finding of this study is that DAPK1 and STAT3 negatively regulate each other in RCC cell lines; moreover, STAT3 promotes miR-93 to suppress the pro-apoptotic and anti-proliferative functions of DAPK1. These findings might reveal a novel pathway between tumorigenesis and inflammatory reactions. Moreover, they offer a new therapeutic perspective for the treatment of RCC.

Conclusions
In summary, we have revealed that miR-93-3p acts as an oncogene in RCC, enhances cell proliferation, and suppresses apoptosis. miR-93-3p was directly regulated by STAT3 and inhibited DAPK1 protein expression. In addition, DAPK1 inactivated STAT3, and STAT3 decreased DAPK1expression. We hope to conduct a new study on the involvement of the STAT3/miR-93-3p/DAPK1 signalling feedback pathway in inflammatory reactions and cancer. The disruption of this pathway might be a promising therapeutic approach in the treatment of RCC.