Dexmedetomidine promotes colorectal cancer progression via Piwil2 signaling

Abstract

Purpose

α2-adrenoceptor agonist dexmedetomidine (DEX) has been reported to promote tumorigenesis. Stem-cell protein Piwil2 is associated with cancer progression. Whether Piwil2 plays a role in tumor-promoting effects of DEX is unknown.

Methods

We examined the expression of Piwil2 in human colorectal cancer (CRC) cell lines with/without DEX treatment. We also studied the roles of Piwil2 in proliferation, invasion, migration, as well as expressions of epithelial-mesenchymal transition (EMT)-related proteins in DEX-treated in vitro and in vivo CRC models. And the experiments with genetic and pharmacological treatments were conducted to investigate the underlying molecular mechanism.

Results

RNA-sequencing (RNA-seq) analysis found Piwil2 is one of most upregulated genes upon DEX treatment in CRC cells. Furthermore, Piwil2 protein levels significantly increased in DEX-treated CRC cancer cells, which promoted proliferation, invasion, and migration in both CRC cell lines and human tumor xenografts model. Mechanistically, DEX increased nuclear factor E2-related factor 2 (Nrf2) expression, which enhanced Piwil2 transcription via binding to its promoter. Furthermore, in vitro experiments with Piwil2 knockdown or Siah2 inhibition indicated that DEX promoted EMT process and tumorigenesis through Siah2/PHD3/HIF1α pathway. The experiments with another α2-adrenoceptor agonist Brimonidine and antagonists yohimbine and atipamezole also suggested the role of Piwil2 signaling in tumor-promoting effects via an α2 adrenoceptor-dependent manner.

Conclusion

DEX promotes CRC progression may via activating α2 adrenoceptor-dependent Nrf2/Piwil2/Siah2 pathway and thus EMT process. Our work provides a novel insight into the mechanism underlying tumor-promoting effects of α2-adrenoceptor agonists.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13402-024-00944-8.

Introduction

Dexmedetomidine (DEX), a highly selective α₂-adrenoceptor agonist, is commonly used during cancer surgeries for both sedation and analgesia. Recent clinical data suggest that perioperative use of DEX is associated with cancer metastasis and recurrence in surgical patients undergoing cancer resection, leading to their undesirable survival outcomes [1, 2], however, other studies did not find a such association [3, 4]. Given these studies are retrospective and have small sample size, the role of DEX in survival outcomes of cancer patients remains controversial. Nevertheless, preclinical studies have indicated that DEX contributes to the growth and metastatic potential of breast, lung, gastric, and colorectal cancer (CRC) [5–10]. Therefore, it has aroused our interest to further investigate the role of DEX in tumorigenesis and explore possible molecular mechanisms.

To unravel the effect of DEX on the cancer biology, we performed an RNA-sequencing analysis, and observed that the stem-cell protein P-element Induced WImpy protein-like RNA-mediated gene silencing 2 (Piwil2) was one of most upregulated genes upon DEX treatment in CRC cells. Initially Piwil2 is considered to play crucial roles in spermatogenesis and embryogenesis [11], recent studies have reported an association between Piwil2 and malignant progression in a various tumors, including gastric, breast, cervical cancers, as well as CRC [12–14]. This suggests that Piwil2 may also play an important role in tumorigenesis. In CRC, higher Piwil2 expression is significantly associated with more aggressive clinical and pathological profiles, leading to poorer 5-year metastasis-free and overall survival [15]. However, whether DEX drives CRC progression via Piwil2 is not yet reported so far.

A recent study has shown that Piwil2 promotes cell proliferation and suppresses cell apoptosis in both cervical cancer and hepatocellular carcinoma cells through competitive association with E3 ubiquitin ligase seven in absentia homologue 2 (Siah2) [16]. Siah2, a known regulator of hypoxia-activated pathways, plays a role in regulating cell growth, proliferation, and differentiation, thereby supporting its function as a tumor promoter in breast, gastric, pancreatic, lung cancers and CRC [17]. However, whether and how DEX drives CRC progression via Piwil2-Siah2 pathway is unknown.

In the present study, we hypothesize that DEX activates the Piwil2-mediated molecular signaling to promote cancer progression. We utilized two human CRC cell lines and xenograft models to investigate the role of Piwil2 in DEX-induced tumorigenesis and its underlying molecular mechanism.

Materials and methods

Antibodies and regents

The antibodies used for western blotting and immunohistochemistry were as follows: anti-Piwil2 (1:1000), anti-Siah2 (1:1000), anti-PHD3 (1:1000), anti-Nrf2 (1:1000), anti-GAPDH (1:10,000), anti-E-cadherin (1:1000), anti-Vimentin (1:1000), anti-N-cadherin (1:1000) and anti-Occludin (1:1000) from ProteinTech (Chicago, USA); and anti-HIF1α (1:500) from Cell Signaling Technology (Danvers, MA, USA); anti-Ki67 (1:500) from Santa Cruz Biotechnology Inc. (Dallas, TX, USA). Dexmedetomidine from Selleck Chemicals (Houston, TX, USA), Brimondine, Yohimbine, Atipamezole, Vitamin K3 and ML385 from MCE (Sovizzo, Italy).

The cancer genome atlas (TCGA) analysis and prognosis analysis of CRC patients

To investigate the role of Piwil2 in CRC progression and prognosis, we analysed the expression profile of Piwil2 in CRC using The Cancer Genome Atlas Colon Adenocarcinoma (TCGA-COAD) and Genotype-Tissue Expression (GTEx) databases through the Gene Expression Profiling Interactive Analysis (http://gepia2.cancer-pku.cn/#analysis). Subsequently, we evaluated the relationship between Piwil2 expression and patient survival outcomes. Kaplan-Meier survival curve of the GSE39582 dataset was used to analyze the overall survival (OS) and disease-free survival (DFS) associated with Piwil2 expression.

Clinical sample collection

CRC tissues from 10 surgical patients (male/female: 5/5) and their matched adjacent non-cancerous tissues were obtained for western blotting and immunohistochemical staining at the Fudan University Shanghai Cancer Center (FUSCC) from February to March 2024. All specimens were acquired after obtaining written informed consent, provided by the Ethics Committee of the FUSCC. The tumour stage was determined according to the American Joint Committee on Cancer (AJCC) Cancer Staging system. The acquisition of CRC tissues specimens was approved by the Ethics Committee of FUSCC (Approval number: 2311286-2).

Screening of CRC cell lines

We obtained the necessary transcriptome data from the Gene Expression Omnibus (GEO) database under the accession number GSE97023 [18]. By employing R packages specifically designed for differential gene expression analysis, we performed statistical analysis to assess whether there were significant alterations in Piwil2 expression levels among CRC cell lines. Based on the outcomes of this analysis, we evaluated the differential expression patterns of Piwil2 in CRC cell lines.

Cell cultures and interventions

Two human CRC cell lines RKO and SW480 were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cancer cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen, USA) and incubated at 37 °C in a 5% CO₂ incubator. All media were supplemented with 10% fetal bovine serum (FBS; Gibco, USA) and 1% penicillin/streptomycin (Invitrogen, USA). Based on concentration-effect and time-effect experiments and previous in vitro cell study [19], the exposure concentration and duration of DEX used in cell studies were selected. α2-adrenergic antagonist Yohimbine (1 μM) and Atipamezole (1 μM) or Siah2 receptor antagonist Vitamin K3 (1 and 10 μM) or Nrf2 inhibitor ML385 (1 μM) was used for pre-treatment for 1 h followed by DEX administration to CRC cells, and effect of another α2-adrenergic agonist Brimondine (1 μM) was investigated similar to DEX experiments.

RNA-sequencing (RNA-seq) analysis

Total RNA was isolated from RKO cells using TRIzol reagent (Invitrogen, USA) for RNA-seq analysis on BGISEQ-500 platform. The P-value was adjusted using the Benjamini-Hochberg false discovery rate method (R package DEseq2). Genes with an adjusted P-value < 0.05, and | log 2 (fold change) | ≥ 1 were deemed to be differentially expressed.

ShRNA transfection

shRNAs for Piwil2 were designed and synthesized by GeneChem Co., Ltd. (Shanghai, China). Full-length Piwil2 was amplified using PCR and subcloned into constructing pcDNA3.1 (Invitrogen, Carlsbad, CA, USA). The target sequences were human Piwil2-shRNA#1, 5′-ACCGGCUGGGUUGAACUAAA-3′; Piwil2-shRNA#2, 5′-ACAGCCUCAAACACUGUGCAC-3′; and negative-control shRNA (sh-NC), 5′-GUACCGCAC GUCAUUCGUAUC-3′. RKO and SW480 cells were transfected with Piwil2-shRNAs or negative-control shRNA using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. Transfection efficiency was assessed by real-time quantitative PCR (qPCR) and western blotting.

Western blotting

CRC cells were lysed with lysis buffer (Beyotime, China), and proteins were extracted following standard protocols. Proteins were separated using 10% SDS-PAGE and transferred to a polyvinylidene fluoride (PVDF) membrane. The membrane was blocked with 5% skim milk in PBST for 1 h at room temperature. The membrane was incubated with primary antibodies overnight at 4 °C, and then was incubated with peroxidase-conjugated secondary antibodies for 2 h at room temperature. Protein bands were detected using a chemiluminescence imaging system (Tanon-5200, Shanghai, China) and quantified using ImageJ software (v1.37, NIH, USA).

Hematoxylin-Eosin (H&E) staining

The H&E staining was performed on the CRC tumor tissues or liver tissues obtained from xenograft models. The paraffin sections of tissue were deparaffinized and rehydrated. Stained with hematoxylin for 10 min, followed by differentiated with 1% hydrochloric acid alcohol for 5 s, and then rinsed with tap water, the sections were placed in warm water at 50 °C to blue. After 3 min incubation in 85% ethanol, the sections were stained with eosin for 3 min. Rinsed for 5 s with tap water, the sections then were dehydrated in gradient ethanol. After sealing the sections with neutral balsam, we utilized a digital section scanner (KF-PRO-005, Ningbo Jiangfeng Bio-information Technology Co., Ltd., China) to acquire images at 100× magnification.

Immunohistochemical staining

Tumor tissues were embedded in paraffin and cut into 4 μm sections, followed by deparaffinization and rehydration. Immunohistochemical staining was performed with antibodies at 4 °C overnight. The sections were incubated with secondary antibodies for 2 h at 37 °C and were mounted in a mounting medium containing glycerol (Beyotime, China). Positively stained tumor cells were analyzed, and images were captured under a light microscope.

Immunofluorescence staining

CRC cells were seeded onto glass slides, fixed with 4% paraformaldehyde solution for 10 min, and permeabilised with 0.3% Triton-X in PBS for 30 min at room temperature. After washed three times in PBS, the cells were blocked with 5% BAS for 30 min to block nonspecific binding sites. Primary antibodies against Ki67 (1:200), E-cadherin (1:200), Vimentin (1:100), Piwil2 (1:200), Siah2 (1: 500) and Nrf2 (1:200) were incubated with the fixed cells at 4 °C overnight, respectively. Next, they were incubated with fluorescence-labelled secondary antibody at 37 °C for 1 h. Finally, DAPI was used to stain nuclei for 10 min. Images were captured by confocal fluorescence microscopy.

Wound-healing assay

CRC cells were cultured in six-well plates overnight. A sterile P200 pipette tip was used to create a straight scratch in the monolayer. After two washes with PBS, the cells were incubated with fresh complete serum-free DMEM containing 1 μM DEX for 12 h, and then cells were cultured in serum-free DMEM. The plates were imaged at 0 and 48 h using an inverted microscope.

Transwell assay

Cell invasion assays were performed in a Transwell chamber (24-well, 8 μm pore size; Corning). CRC cells were re-suspended in serum-free DMEM medium, and 5 × 10⁴ cells suspended in 200 μl were added to the upper chamber membranes. The bottom was coated with 1 mg/ml BD Matrigel matrix, and 800 μl of complete medium was placed into the lower chamber as a chemoattractant. The cells were then incubated at 37 °C in a 5% CO₂ incubator. After 48 h, the non-invading cells on the membrane surface were removed using a cotton swab. The invading cells on the lower surface were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet for 30 min. The number of invading cells was counted in four fields for each transwell, and the data were analyzed using ImageJ software.

Coimmunoprecipitation (Co-IP)

Cell protein lysates were separated using lysis solution, which was mixed with primary antibodies against Piwil2, Nrf2, Siah2, and IgG, and incubated overnight at 4 °C for endogenous IP. Next, the lysates were incubated with 50 μl protein Dynabeads (Thermo, MA, USA) at 4 °C for 6 h. The beads were washed 3 times with RIPA buffer to remove impurities. Finally, 20 µL of IP lysate was added to 2 × load buffer and boiled. The results were analysed by western blotting.

Chromatin immunoprecipitation (Ch-IP) assay

A Ch-IP detection kit (Cell Signaling Technology, MA, USA) was used. Briefly, cells (1.0 × 10⁷) were treated with 1% formaldehyde. Glycine (2.5 M) was added to prevent cross-linking. The chromatin was examined ten times using ultrasound. After centrifugation, the supernatant was incubated with anti-Piwil2 antibody, anti-Nrf2 antibody and IgG at 4 °C. The agarose beads were subjected to immunoprecipitation. DNA-protein cross-links were reversed in a 65 °C water bath for 6 h. The enriched sequences of the purified DNA were analysed by PCR. The oligonucleotide primer sequences for Piwil2 were 5′-ACTGGCTCAAGGTGGAAGGA-3′ forward and 5′-CAGGCTGCCTCAGTTCATGG-3′ reverse. Additionally, a re-chip assay was used to verify whether Nrf2 occupied the same binding site in the Piwil2 promoter region. Chromatin in beads was eluted with 10 mM DTT after a standard Ch-IP procedure. The eluent was diluted with ultrasound buffer prior to Ch-IP treatment.

Real-time quantitative PCR (qPCR)

Total RNA was extracted from RKO and SW480 cells using lysis and wash buffer (Invitrogen, Carlsbad, CA) for qPCR analysis. Subsequently, cDNA synthesis was conducted with a reverse transcription kit (Invitrogen) to convert RNA into cDNA templates. Specific primers were designed for the target genes Piwil2 (forward: 5′-ACTGGCTCAAGGTGGAAGGA-3′, reverse: 5′-CAGGCTGCCTCAGTTCATGG-3′) and reference gene GAPDH (forward: 5′-CGCTCTCTGCTCCTCCTGTT-3′, reverse: 5′-CCATGGTGTCTGAGCGATGT-3′), and qPCR reactions were prepared with SYBR Green master mix (A0012-R2, EZBioscience, USA). The qPCR reactions were performed in triplicates on a thermal cycler with cycling conditions set at 95 ℃ for 3 min, followed by 40 cycles at 95 °C for 12s and 55 °C for 40 s. Data analysis involved utilizing specialized software to calculate Ct values and determine gene expression levels relative to the reference gene. Each trial was independently replicated to ensure the reproducibility and reliability of the results.

Animal experiments

Male BALB/c nude mice aged 4–6 weeks were used for the in vivo study. Mice were obtained from Shanghai SLAC Laboratory Animal Co. (Shanghai, China) and housed in a controlled room with constant temperature and access to a natural day/night cycle of food and water. The protocol of animal study was approved by the Animal Ethics Committee of Fudan University (Approval number: 202210023Z). All animals were treated in accordance with the guidelines for the Care and Use of Laboratory Animals. The nude mice were randomly divided into four groups (n = 6/group): sh-NC, sh-Piwil2, sh-NC + DEX, and sh-Piwill2 + DEX. For the mice in sh-NC + DEX and sh-Piwil2 + DEX groups, 25 μg/kg DEX was injected intraperitoneally for 7 days after cancer cells were inoculated, as previous study described [7, 20], while same volume of normal saline was administrated to mice in other two groups.

Subcutaneous tumor model

According to xenograft model used in a previous study [7], control RKO cells or lentivirus-mediated stable Piwil2-knockdown RKO cells (200 μl, 1 × 10⁶ cells/ml) were injected subcutaneously into the nude mice, and tumor volume was measured daily using the following formula: volume = (length × width²)/2. The nude mice were sacrificed 21 days after the administration of DEX, the tumors were excised, weighed, and imaged followed by immunohistochemical analysis.

Liver metastasis model

A hemisplenectomy model of liver metastasis was constructed as previous described [21]. Briefly, a laparotomy was performed by using left subcostal incision in the nude mice anesthetized with 80 mg/kg ketamine intraperitoneal injection. After splenic blood vessels located at the inferior end of the spleen, the spleen was divided at its center carefully, and its upper pole was placed back into the peritoneum to avoid contamination. Then 200 μl (1 × 10⁶ cells/ml) single-cell suspension of control RKO cells or lentivirus-mediated stable Piwil2-knockdown RKO cells were inoculated slowly into the inferior end of the spleen. Five minutes later after CRC cells transplantation, the exposed hemispleen was removed. The abdominal wall was closed and local bupivacaine was used for postoperative analgesia, and 300 mg/kg cefazolin was administrated intraperitoneally for infection prevention. The mice livers were harvested 7 days after live imaging, and the liver metastatic nodules were counted with visual inspection.

Small animal intravital imaging

The live imaging was performed at the 21st day after establishment of liver metastasis model. Briefly, the mice were intraperitoneally injected with 150 mg/kg of fluorescein, and 10 min later, they were anesthetized with isoflurane and scanned with small animal PET (FX PRO, Bruker, MA, USA) with appropriate position. The liver metastatic lesions were analyzed using a Small Animal Live Imaging System (Bruker, MA, USA).

Statistical analysis

Data are expressed as the mean ± SEM of at least three experiments. All statistical analyses were performed using GraphPad Prism 9.0 (GraphPad Software Inc., CA, USA). An independent sample t-test was used to compare individual data with control values, and a one-way analysis of variance was used to compare the data of multiple groups. Analysis of variance was used to compare the differences among the groups under different conditions. P < 0.05 was considered statistically significant.

Overall, our experimental procedure is depicted in Fig. 1a and comprises three primary components. Firstly, we analyzed the expression profile of Piwil2 in CRC using the databases and evaluated the relationship between Piwil2 expressions and patient outcomes. Secondly, we established an in vitro model using CRC cells treated with dexmedetomidine to investigate the mechanisms underlying its impact on CRC through a series of molecular biological experiments. Thirdly, we developed in vivo models that involved subcutaneous tumor-bearing CRC and liver metastasis to evaluate the effects of DEX on the proliferation and metastasis of CRC.

Fig. 1.

Schematic workflow and the correlation between high Piwil2 expression in colorectal cancer (CRC) and clinical prognosis. a Schematic workflow depicting the influence of dexmedetomidine (DEX) on the malignant progression of CRC; b The expression of Piwil2 in human CRC tissues (n = 290) and normal tissues (n = 349) analyzed from the combined TCGA and GTEx databases; c, d Overall survival (c) and diseases-free survival (d) were compared between patients with high and low Piwil2 expression in GSE39582 databases. e The protein expressions of Piwil2 in human tumor tissues and adjacent normal tissues measured by western blot (n = 10); f Quantification of the protein expression of Piwil2 (e); g Piwil2 expression in tumor tissues and adjacent normal tissues measured by immunohistochemical staining, scale bar = 50 μm; h Quantification of the expressions of Piwil2 (g). All of the data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001

Results

Higher Piwil2 expression associated with poorer prognosis in CRC patients

We analyzed the expression profile of Piwil2 in CRC using the databases, and the results showed that the expressions of Piwil2 were significantly increased in CRC tissue samples than in normal tissue ones (P = 0.000, Fig. 1b). Furthermore, we evaluated the relationship between Piwil2 expressions and patient outcomes. The Kaplan-Meier survival curve analysis of the GSE39582 dataset showed that higher Piwil2 expression in CRC was associated with lower OS and DFS rate (Fig. 1c, d, P < 0.05). To validate the above results, we performed western blotting on CRC specimens of 10 patients obtained surgically at FUSCC. The results results indicated that the Piwil2 expressions were significantly higher in CRC tissues compared to adjacent normal tissues (Fig. 1e, f, P < 0.001). Moreover, immunohistochemical analysis further confirmed these observations (Fig. 1g, h, P = 0.0006). The collective findings presented here provide compelling evidence for the upregulated expression levels of Piwil2 in colon cancer tissues and its correlation with an unfavorable prognosis.

Dexmedetomidine up-regulates Piwil2 expression and promotes tumorigenesis in CRC cell lines

The RNA-seq analysis showed in DEX-treated RKO cells, 23 differentially expressed genes (DEGs) were found, of which 18 genes up-regulated and 5 genes down-regulated when compared with control cells. Among those DEGs, Piwil2 expression showed a 2.8-fold up-regulation in DEX-treated cells (Fig. 2a, b). According to the expression profiles of Piwil2 presented in Fig. 2c, we identified 6 cell lines (SW480, RKO, SW620, HT29, LOVO and DLD1) that exhibited alterations in Piwil2 expression. Notably, the RKO and SW480 cell lines showed a significant increase in Piwil2. Therefore, we selected RKO and SW480 cell lines for subsequent in vitro investigations. Furthermore, we investigated the effects of different concentrations of DEX exposure on Piwil2 expression. In RKO cells, 0.1–1 μM DEX could significantly up-regulate Piwil2 expression; while in SW480 cells, only 1 μM DEX significantly up-regulated Piwil2 expression (Supplementary Fig. 1a, b). Additionally, we assessed the impact of DEX exposure duration on Piwil2 expression in both cell lines. The results indicated that Piwil2 expression was significantly up-regulated after 12 h and 24 h of exposure to 1 μM DEX, but not after a 6 h-exposure (Supplementary Fig. 1c, d). Therefore, we used 1 μM DEX to treat CRC cells for 12 h in the following in vitro studies. Actually, 1 μM DEX treatment for 12 h increased the Piwil2 protein (RKO: P = 0.0017 and SW480: P = 0.0015; Fig. 2d, e) expressions in both cancer cell lines.

Fig. 2.

Dexmedetomidine (DEX) increases Piwil2 expression and promotes proliferation, migration and invasion of CRC cells. a Volcano plot of differentially expressed genes from RNA-seq analysis of RKO lineage cells, upregulated (red, log 2 fold change > 2; FDR < 0.05); b Heat map of data from a; c Schematic diagram of Piwil2 expression in CRC cell lines; d Piwil2 expression in RKO and SW480 cells after DEX treatment measured by western blot (n = 3/group); e Quantification of Piwil2 protein expressions in two CRC cell lines (d); f, g Real-time PCR indicated that mRNA was down-regulated by interference of Piwil2. h Piwil2 expressions in RKO and SW480 cells grouped by sh-NC, DEX + sh-NC, and DEX + sh-Piwil2, respectively, measured by western blot (n = 3/group); i Quantification of Piwil2 protein expressions in two CRC cell lines (h); j Immunofluorescence staining of Ki67 (green) and DAPI (blue) in two CRC cell lines (n = 5/group), scale bar = 50 μm; k Quantification of the relative expression of Ki67; l, m Wound healing assay was used to measure the migration of CRC cells. The wound space test was photographed at 0 and 48 h after treatments (n = 5/group), scale bar = 50 μm; n, o Transwell test was conducted to analyze the invasion of CRC cell lines (n = 5/group), scale bar = 100 μm. All of the data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001

Moreover, sh-piwil2 was utilized to knockdown Piwil2 expression in two CRC cell lines. Firstly, we assessed the knockdown efficiency of sh-Piwil2 on Piwil2 expression at the mRNA level. The results demonstrated a significant decrease in Piwil2 mRNA expression in both the sh-Piwil2#1 and sh-Piwil2#2 groups compared to the sh-NC group (P = 0.0000 and P = 0.0002 respectively; Fig. 2f). Furthermore, treatment with DEX resulted in a notable upregulation of Piwil2 mRNA expression, which was subsequently reduced after co-incubation with sh-Piwil2#1 and sh-Piwil2#2 (P = 0.0004, P = 0.0001 and P = 0.0003 respectively; Fig. 2g). To validate these results, we further examined Piwil2 expression at the protein level (Fig. 2h, i), DEX administration significantly increased Piwil2 protein expression, while knockdown of Piwil2 led to a significant reduction in Piwil2 protein expression. Subsequent cell biology studies were carried out to evaluate the influences of Piwil2 knockdown on the proliferation, migration, and invasion capacities of RKO and SW480 cell lines. In comparison to the sh-NC group, DEX notably enhanced cell proliferation, as indicated by an increase in Ki67 expression; However, Piwil2 knockdown inhibited the proliferative effect of DEX (Fig. 2j, k). Similarly, DEX-treated cells exhibited accelerated wound closure and enhanced migration ability when compared with control cells (Fig. 2l–o). Notably, invasion and metastasis were significantly attenuated in DEX-treated Piwil2-knockdown CRC cells compared to wild-type CRC cells. Collectively, our findings suggest that DEX promotes the proliferation, invasion, and metastasis of CRC cells by upregulating Piwil2 expression.

Dexmedetomidine promotes tumorigenesis by activating Piwil2/Siah2/HIF1α pathway

To answer whether Piwil2-Siah2 signaling mediates DEX tumor-promoting effects, a Co-IP analysis was performed, and the results showed an interaction between Piwil2 and Siah2 (Fig. 3a). Furthermore, Piwil2 knockdown significantly reduced the expression of Siah2 and HIF1α, while increasing PHD3 expression (Fig. 3b–d). Similarly, Siah2 inhibitor Vitamin K3 treatment significantly decreased HIF1α expression and increased PHD3 expression, while without affecting Piwil2 levels (Fig. 3e–g). The immunofluorescence experiment also showed Piwil2 knockdown reversed DEX-induced Siah2 overexpressions (Fig. 3h). These data suggested that Siah2 act as a substrate of Piwil2, and Piwil2 activates the Siah2/PHD3/HIF1α pathway through its binding to Siah2. Furthermore, Vitamin K3 inhibited proliferation (Supplementary Fig. 2a, b), migration (Supplementary Fig. 2c, d) and invasion (Supplementary Fig. 2e, f) in both DEX-treated CRC cell lines. Therefore, our findings further suggest that DEX promotes tumorigenesis through the involvement of Piwil2 as a central molecule associated with a specific signaling pathway.

Fig. 3.

Dexmedetomidine (DEX) activates Piwil2-mediated Siah2/PHD3/HIF1α pathway. a Co-IP experiment showed the interaction between Piwil2 and Siah2 in RKO cells; b The effect of Piwil2 knockdown on the expressions of Siah2, PHD3 and HIF1α detected by western blot in DEX-treated RKO (left) and SW480 (right) cells (n = 3/group); c, d Quantification of the expressions of Siah2, PHD3 and HIF1α in RKO and SW480 cells; e The effect of Vitamin K3 (a Siah2 inhibitor) on the protein expressions of Piwil2, Siah2, PHD3 and HIF1α detected by western blot in DEX-treated RKO (left) and SW480 (right) cells (n = 3/group); f, g Quantification of the expressions of Piwil2, Siah2, PHD3 and HIF1α in RKO and SW480 cells; h DEX increased both Piwil2 and Siah2 expressions while Piwil2 knockdown decreased Siah2 expression detected by immunofluorescence in both CRC cell lines (n = 5/group), scale bar = 50 μm. All of the data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. Con: Control; VK3: Vitamin K3

Dexmedetomidine increases epithelial-mesenchymal transition (EMT) through up-regulating Piwil2 expression

To investigate whether Piwil2 mediates the effects of DEX on EMT, a biological process involved in cancer invasion and metastasis, we assessed the expressions of EMT-associated proteins in Piwil2 and wild-type CRC cells. The results showed that expressions of E-cadherin and occludin significantly decreased, while N-cadherin and vimentin expressions increased significantly in DEX-treated CRC cells (Fig. 4a–c). Notably, Piwil2 knockdown reversed the DEX-induced changes in EMT protein markers. Furthermore, confocal imaging also observed the altered expressions of E-cadherin and vimentin (Fig. 4d). These results suggest that DEX promotes the EMT process by upregulating Piwil2 expression.

Fig. 4.

Dexmedetomidine (DEX) activates Piwil2-mediated EMT process in CRC cell lines. a The effect of dexmedetomidine and Piwil2 knockdown on the expressions of EMT related proteins detected by western blot in RKO (left) and SW480 (right) cells (n = 3/group); b, c Quantification of the expressions of E-cad, Vim, Occ, and N-cad in RKO and SW480 cells; d Immunofluorescence staining of E-cadherin (red),vimentin (green) and DAPI (blue) in CRC cell lines treated with DEX and Piwil2 knockdown (n = 5/group). All of the data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. EMT: mesenchymal-epithelial-transition; E-cad: E-cadherin; Vim: Vimentin; Occ: Occludin: N-cad: N-cadherin

Dexmedetomidine activates Piwil2 signaling dependent on α2 adrenoceptor

To explore whether Piwil2-mediated pro-tumor effect by DEX depends on activation of α2 adrenoceptor, we pre-incubated CRC cell lines with both α2 adrenoceptor inhibitor yohimbine (1 μM) and atipamezole (1 μM) followed by DEX treatment. The results showed that yohimbine significantly down-regulated DEX-induced increases in the expressions of Piwil2, Siah2 and HIF1α, while up-regulating the expression of PHD3 (Supplementary Fig. 3a, b). Additionally, yohimbine dramatically reversed DEX-induced proliferation, invasion, and migration (Supplementary Fig. 3c–h). Similarly, treatment with 1 μM atipamezole also down-regulated DEX-induced increased protein expressions of Piwil2, Siah2 and HIF1α, and up-regulating PHD3 protein expression (Supplementary Fig. 3i, j). Therefore, these data indicate that Piwil2/Siah2/PHD3/HIF1α axis promotes CRC progression may depending on activation of α2 adrenoceptor by DEX.

To further investigate whether Piwil2 expression is mediated by activation of α2 adrenoceptor, we treated CRC cells with another α2 adrenoceptor agonist, Brimonidine. The results showed that like the findings in DEX experiments, Brimonidine treatment also enhanced Piwil2 expressions in both CRC cell lines (Supplementary Fig. 4a, b). Similarly, Brimonidine also increased the expression of N-cadherin and vimentin, while decreasing the expressions of E-cadherin and occludin (Supplementary Fig. 4a, b). Additionally, compared to control cells, Brimonidine significantly increased the number of Ki67 positive cells (Supplementary Fig. 4c, d), accelerated wound healing (Supplementary Fig. 4e, f), and promoted cell invasion (Supplementary Fig. 4g, h) in both CRC cell lines. Collectively, these findings suggest that α2 adrenoceptor agonists promote CRC progression by activating Piwil2 signaling.

Piwil2 is upregulated by dexmedetomidine via Nrf2

Since Piwil2 was up-regulated at mRNA level, it is assumed that DEX changes Piwil2 expression very likely involving a transcription factor. The transcription factor Nrf2 has been reported to directly target Piwil2 in the development of radiation-induced lung injury [22]. Interestingly, Nrf2 is also a key transcription factor implicated in the tumor progression [23], Therefore, we investigated the role of Nrf2 in DEX-induced increase in Piwil2 expression. The results showed that DEX significantly increased Nrf2 expression in CRC cells (Fig. 5a, b). Moreover, Yohimbine reversed DEX-induced up-regulation of Nrf2 expression (Fig. 5c, d), suggesting an α2 receptor-dependent effect. Co-IP assay demonstrated an interaction between Nrf2 and Piwil2 in CRC cells, and DEX enhanced the binding of Nrf2 to Piwil2 (Fig. 5e). The further Ch-IP analysis found that Nrf2 targeted the Piwil2 promoter to regulate its transcriptional activity (Fig. 5f–h). Three potential Nrf2 binding domains were identified (Fig. 5g), and Nrf2 was found to specially bind to primer-1 domain of Piwil2 gene. The Nrf2 antagonist ML385 (1 μM) significantly reduced DEX-induced increased Piwil2 expression in CRC cells (Fig. 5i–k). Moreover, immunofluorescence results showed that co-labeled Nrf2 and Piwil2 in the cytoplasm before DEX exposure, while the staining of Nrf2 and Piwil2 increased in both cytoplasm and nucleus after DEX treatment (Fig. 5l), suggesting their intracellular co-localization, and our results further confirmed DEX upregulated Piwil2 expression via modulating Nrf2 expression.

Fig. 5.

Dexmedetomidine (DEX) induces increases in Nrf2 expression contributing to up-regulation of Piwil2 in CRC cells. a Nrf2 expression after DEX treatment measured by western blot in RKO (up) and SW480 (down) cells (n = 3/group); b Quantification of Nrf2 protein expressions in two CRC cell lines (a); c The effects of α2 adrenoceptor inhibition (Yohimbine) on expression of Nrf2 detected by Western blot in DEX-treated RKO (up) and SW480 (down) cells (n = 3/group); d Quantification of Nrf2 expressions in CRC cell lines (c); e The interaction between Nrf2 and Piwil2 after DEX treatment in RKO cells detected by Co-IP test; f The enriched consensus motif associated with binding sites of transcription factor Nrf2 in Piwil2 gene; g The predicted possible binding sites in the primer-1 domain of Piwil2 gene by Jaspar database; h ChIP results confirmed that Nrf2 promoted Piwil2 transcription through binding to its primer-1 domain; i–k The effect of ML385 (a Nrf2 inhibitor) on the Piwil2 protein expressions in DEX-treated CRC cells (n = 3/group); l Immunofluorescence staining of Piwil2 and Nrf2 after ML385 treatment in DEX-treated CRC cells. scale bar = 50 μm. All of the data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. Con: Control; Y: Yohimbine

Dexmedetomidine promotes Piwil2-mediated tumor progression in xenograft models

In the vivo experiments, tumor samples obtained from subcutaneous xenograft mice treated with sh-NC + DEX were found to be larger than in size compared to those in sh-NC group (Fig. 6a). However, the tumor volumes (Fig. 6b) and weights (Fig. 6c) in the sh-Piwil2 + DEX group were lower than those in the sh-NC + DEX group. Additionally, immunohistochemical tests further showed that subcutaneous tumors extracted from the sh-Piwil2 + DEX group exhibited lower levels of Siah2 and Ki67, and higher levels of E-cadherin (Fig. 6d–g). These findings suggest that DEX promotes in vivo tumor growth may through the activation of Piwil2 signaling.

Fig. 6.

Dexmedetomidine (DEX) promotes Piwil2-meidated proliferation and metastasis in animal models. a Representative image of the human tumor xenografts removed from nude mice 28 days after RKO cells subcutaneous implantation (n = 6/group); b The tumor growth curves in nude mice with/without DEX treatment, subcutaneously implanted with wild-type and Piwil2-knockdown RKO cells; c The tumor weight in nude mice with/without DEX treatment, subcutaneously implanted with wild-type and Piwil2-knockdown RKO cells; d Representative images of H&E staining and immunohistochemical staining for Siah2, E-cadherin and Ki67 in harvested xenografted tumor samples. H&E: 100×, scale bar = 50 μm; e–g Quantification of the expressions of Siah2, E-cadherin and Ki67 in harvested xenografted tumor samples (n = 6/group); h Development of the liver metastasis model. RKO cells were injected into the inferior pole of the spleen (up), and then the inferior pole of spleen was removed along with its blood vessels (down); i, j Representative images of small animal live imaging 21st day (i) and liver samples harvested 28th day (j) after injection of wild-type or Piwil2 knockdown RKO cells into spleen (n = 6/group); the red arrows represent metastatic nodules in the livers; k H&E staining for metastatic colonies and immunohistochemical staining for E-cadherin and Ki67 expressions in harvested mice liver metastatic samples. H&E: 100×, scale bar = 50 μm; l Quantification of the expressions of E-cadherin and Ki67 in harvested mice liver metastatic samples (n = 6/group). All of the data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001

Furthermore, in the splenic hemiplectomy model (Fig. 6h), small animal live imaging (Fig. 6i) at postoperative 21st day showed the mice receiving RKO cell implantation and DEX treatment (sh-NC + DEX group) had more liver metastatic colonies than those receiving sh-Piwil2 RKO cell implantation and DEX treatment (sh-Piwill2 + DEX group). Similarly, at the postoperative 28th day, more metastatic colonies were observed in the livers of sh-NC + DEX-treated mice than in the sh-NC group. However, animals receiving sh-Piwil2 CRC cells transplantation alone exhibited fewer liver metastatic colonies (Fig. 6j, k). Further immunohistochemical analysis of liver tissue demonstrated that the sh-Piwil2 + DEX group exhibited higher E-cadherin staining and lower Ki67 staining than the sh-NC + DEX group (Fig. 6k, l). Overall, these in vivo results are consistent with the in vitro findings, indicating that Piwil2 may mediate DEX-induced CRC progression.

Discussion

Previous experimental studies have shown DEX could enhance cancer cell proliferation and migration via an antiapoptotic mechanism [21]. The present study confirmed that DEX promoted tumorigenesis in CRC cell and animal models, and found a novel underlying molecular mechanism, highlighting the role of Piwil2 in DEX-induced cancer progression (Fig. 7).

Fig. 7.

A model diagram of dexmedetomidine tumor-promoting effects in colorectal cancer. Dexmedetomidine increases α2-adrenoceptor-dependent Nrf2 expression by which binds to Piwil2 promoter to increase Piwil2 transcription. In turn, up-regulation of Piwil2 level induces activation of Siah2/PHD3/HIF1α pathway to facilitate EMT process, which promotes tumorigenesis and metastasis of colorectal cancer

Our and other preclinical researches have indicated that DEX promotes tumor growth and metastasis in both in vitro and in vivo models, leading to decreased animal survival [5, 6, 8]. However, there are other studies also reported that DEX could suppress tumor growth in two murine CRC cell lines (MCA38 and CT26) [20] and hepatocellular carcinoma cells [24]. Furthermore, a long-term follow-up clinical study on postoperative delirium showed that perioperative low-dose DEX infusion did not influence the 3-year overall survival [25], nevertheless, most patients in this study were non-cancerous. Recent clinical studies show conflicting results regarding the correlation between administration of DEX and cancer recurrence and overall survival in patients undergoing cancer surgeries [1–4, 26]. Therefore, the potential pro-tumor risk makes perioperative use of DEX controversial in cancer patients. However, directly translating our findings in preclinical models to clinical practice would be imprudent. More clinical trials are required to investigate the relationship between short-term or long-term use of DEX and the outcomes of cancer patients.

The human intestinal mucosal cells express α2-adrenoceptors [27]. Activation of α2-adrenoceptor exhibits a positive effect on the proliferation of various cancer types [9, 10], whereas inhibiting the α2-adrenoceptor can inhibit the proliferation of cancer cells [28]. Consistent with previous study [7], we have demonstrated that α2-adrenoceptor agonists, both DEX and Brimonidine, exhibited pro-tumor effects. On the contrary, both α2-adrenoceptor inhibitors yohimbine and atipamezole abolished the effects of DEX on the CRC cells. Up-regulations of the anti-apoptotic proteins Bcl-2 and Bcl-xL is considered as a molecular mechanism underlying pro-tumor effect by DEX [7]. Our findings provide a novel molecular mechanism involving Piwil2 signaling that mediates the effects of α2-adrenoceptor activation. Nrf2, a transcription factor known for inducing cytoprotective gene, is activated by DEX, leading to its anti-oxidative and anti-inflammatory effects [29]. Nrf2 activation in healthy cells can prevent tumorigenesis, while its constitutive activation in cancer cells promotes tumor growth and metastasis [30] via rewiring metabolism to enhance proliferation [31]. Interestingly, for the first time, our results found that Nrf2 targeted the promoter of Piwil2 gene to increase its expression and their interaction in CRC cells, suggesting Piwil2 may be a downstream target gene of Nrf2.

The cancer/testis antigen protein Piwil2 plays an important role in spermatogenesis, stem cell self-renewal and maintenance, RNA silencing, and translational regulation [13]. The overexpression of Piwil2 can lead to the development of gynaecological cancers [32], breast cancer [33], gastroesophageal cancer [34] and CRC [12], suggesting its association with tumorigenesis. Surprisingly, Piwil2 overexpression shows an inverse association with tumor progression in bladder cancer [35]. In this study, we present novel evidence that DEX significantly promote cancer progression via Piwil2 signaling. These findings imply that DEX may exert diverse biological effects specific to different types of cancer cells.

Our results demonstrate that Piwil2 exerts a direct regulatory influence on Siah2 expression, leading to downregulation of PHD3 expression and upregulation of HIF1α expression. Earlier evidence has suggested the involvement of Siah2-PHD3 axis in regulating HIF1α stability and tumor development [36], whereas the hypoxic response is closely linked with tumorigenesis [37]. Recent studies have shown that PHD3 act as suppressor of tumor growth, conversely, the loss of PHD3 triggers supply of nutrition and oxygen essential for tumor growth through elevating the expression of angiogenic factors [38]. This obviously contributes to the sustained growth of tumor. As a result, our findings confirm that the impact of DEX on CRC relies on the Piwil2-mediated activation of Siah2/PHD3/HIF1α pathway, leading to the progression of EMT, which represents a critical regulatory process involved in CRC invasion and metastasis.

Our study had several limitations. Firstly, we demonstrated the α2 adrenoceptor-dependent Piwil2 expressions in vitro, but we do not examine these in in vivo experiments. Secondly, while DEX is commonly used during perioperative period or intensive care unit for patient sedation. In this study, we administered DEX under non-surgical condition to investigate its pro-tumor effects. Lastly, the clinical therapeutic plasma concentrations of DEX range from 1 to 10 nM [39], and previous studies have used 1 nM DEX to investigate the proliferation and migration of cancer cells [9]. In this study, we used 1 μM DEX to investigate its effects on CRC cell biology in vitro, as some preclinical studies have administrated a concentration of 1 μM DEX in cell experiments [40, 41]. Interestingly, a recent experimental study showed that α2 adrenoceptor agonists clonidine and guanabenz exhibited strong anti-tumor effects through an on-target action exerted on tumor microenvironment, albeit at a tenfold higher dose than previously used [42]. Therefore, despite the likely higher concentration than clinically relevant one, the DEX concentration applied in present in vitro study may be plausible for understanding the on-target action exerted on tumor cells.

In conclusion, we have identified that DEX up-regulates Piwil2 expressions via increasing Nrf2 levels. This, in turn activates the Siah2/PHD3/HIF1α pathway, promoting the EMT process of CRC cells, and consequently enhancing CRC malignancy. These findings provide a novel insight into the pro-tumor role of DEX in preclinical models. However, further research is needed to determine their clinical implications in cancer surgery.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

DEX

Dexmedetomidine

CRC

Colorectal cancer

EMT

Epithelial-mesenchymal transition

Co-IP

Coimmunoprecipitation

Ch-IP

Chromatin immunoprecipitation

Author contributions

J.D., J.C., Y.W., Y.D., X.L. and Z.H. conducted the experiments; J.D., J.C. and Y.W., analyzed and explained the data, J.D. and J.C. prepared the figures. J.Z. designed the experiments; and J.D. and J.Z. wrote the main manuscript text.

Funding

Supported by Nature Science Foundation of China (to Jun Zhang, No. 81971002 and to Jing Dong, No. 81701045) and Shanghai Committee of Science and Technology (No. 21Y11902300).

Data availability

The datasets used and/or analyzed in the current study are available on reasonable request.

Declarations

Summary

α2-adrenoceptor agonist dexmedetomidine has been associated with tumor growth, invasion and metastasis, however, the underlying mechanism is not well known. We found that dexmedetomidine can promote tumorigenesis and metastasis via Piwil2/Siah2/PHD3/HIF-1α pathway-mediated epithelial-mesenchymal transition process, which was dependent on α2-adrenergic receptor activation.

Competing interests

The authors declare no competing interests.