Introduction

Colorectal cancer (CRC) represents a highly prevalent gastrointestinal neoplasm, rated as the second most deadly cancer following lung cancer (Siegel et al. 2020). It was estimated that in the United States, CRC contributed to over 53,000 deaths in 2020 (Siegel et al. 2020). The etiology of CRC begins as low-grade dysplasia, the slow development of which contributes to polyp generation in the colorectal lumen, finally progressing to a tumor with considerable size (Choi et al. 2015). Gender, obesity, a sedentary lifestyle, heavy consumption of alcohol and red meat, and smoking cigarettes are commonly believed as predisposing factors of CRC (Patel et al. 2022). Surgery with adjuvant chemotherapy and radiotherapy is the cornerstone of CRC therapy, the routine application of which fails to overcome tumor recurrence, metastasis as well as drug resistance, however (Simard et al. 2019; Fakih 2015; Ma et al. 2023). Accordingly, the identification of the latent mechanism of CRC may be conducive to the development of novel treatment regimens to increase the lifespan of patients with CRC.

Autophagy is a self-cannibalization process in which biomacromolecules or damaged organelles are phagocytosed and encapsulated to sustain favorable survival conditions through providing cellular energy and protection under the circumstances of destructive environments, such as nutrient deficiency, drug toxicity, and hypoxia (Parzych and Klionsky 2014). Autophagy is implicated in diverse physiopathological processes, including oncogenesis, cellular development, apoptosis and survival (Klionsky et al. 2021). Autophagy-related gene 10 (ATG10) is known as an E2-like activating enzyme essential for the synthesis of the autophagosome complex (Hu and Reggiori 2022). Plenty studies have determined that ATG10 may act as a prognostic biomarker in CRC (Hu et al. 2022) and modulate the chemoresistance and metastasis of CRC cells (Sun et al. 2020; Jo et al. 2017).

DEAD/H box RNA helicases (DDX) are a group of proteins comprised of rapid amplified putative ATP-dependent RNA helicase that are related to tumor initiation and advancement (Zhang et al. 2021). Accumulated reports have well documented DEAD/H box RNA helicase 10 (DDX10) as a tumor promoter in lung adenocarcinoma (Liu et al. 2021), ovarian cancer (Gai et al. 2016) and osteosarcoma (Shi and Hao 2019). Zhou et al. have established that DDX10 mainly exerts oncogenic activities in CRC (Zhou et al. 2022). Recent literature has revealed that DDX10 can be degraded by promoting SQSTM1-mediated selective autophagy (Li et al. 2023). Importantly, ATG genes including ATG3, ATG7 and ATG10 were predicted to be negatively regulated by DDX10 by Biogrid database.

Thence, it is speculated that DDX10 may also participate in the process of CRC through mediating ATG-dependent autophagy, which is the objective of our present work.

Materials and methods

Cell lines, reagents and siRNA transfection

Normal human intestinal epithelial cell line NCM460 (SUNNCELL, Wuhan, China) was hatched in Roswell Park Memorial Institute (RPMI)-1640 medium (Welgene, Gyeongsan, Korea) and CRC cell lines (SW620, DLD-1 and LS174T) supplied by BeNa Culture Collection (Shanghai, China) were all subjected to Dulbecco’s modified Eagle medium (DMEM; Welgene, Gyeongsan, Korea). All mediums were appended with 10% fetal bovine serum (FBS; Bovogen Biologicals, Melbourne, VIC, Australia), along with 1% antibiotics (Welgene, Gyeongsan, Korea) in a water saturated atmosphere equipped with 5% CO2 at 37 °C. Besides, SW620 and DLD-1 cells were pretreated by 1 µM autophagy inhibitor 3-Methyladenine (3-MA; HARVEYBIO, Beijing, China) for 2 h (Huang et al. 2022), and then treated by EZ Trans Lipo transfection reagent (Life-iLab, Shanghai, China) for transfection of siRNAs targeting DDX10 (siRNA-DDX10-1/2) or/and ATG10 (siRNA-ATG10-1/2) as well as the relevant control siRNA (siRNA-NC) (Genomeditech, Shanghai, China).

5-Ethynyl-2’-deoxyuridine (EDU) staining

By means of iClick™ EdU Andy Fluor™ 488 Imaging Kit (GeneCopoeia, Inc.), cell proliferation was appraised in compliance with the indicate protocol. In brief, SW620 and DLD-1 cells (1 × 104 cells/well) prepared in 96-well plates received indicated treatment and transfection, followed by being mixed with 20 µM EDU working solution for 2 h. After being immobilized by 3.7% paraformaldehyde for 15 min and perforated with 0.5% Triton X-100 for 20 min at room temperature, the incorporated EdU was visualized by means of a click reaction using Alexa Fluor 488 azide for half an hour without exposure to light. Images were prepared for observation under a fluorescence microscope (Thermo Fisher Scientific) after the nuclear DNA was dyed by Hoechst 33,342.

Terminal-deoxynucleoitidyl transferase mediated nick end labeling (TUNEL)

Cell apoptosis was appraised applying TUNEL Apoptosis Assay kit (Shanghai Acmec Biochemical Co., Ltd) in agreement with the manufacturer’s instructions. In short, following indicated treatment and transfection, 4% paraformaldehyde was utilized to treat SW620 and DLD-1 cells for immobilization, followed by permeabilization with 0.1% Triton X-100. Subsequently, the cells were cultivated with TUNEL reaction reagent for 1 h and the nuclei were labeled with 10 mg/ml DAPI for 10 min. Finally, five random fields of view were selected to capture images using a fluorescence microscope (Thermo Fisher Scientific).

Sphere formation assay

SW620 and DLD-1 cells (1 × 103 cells per well) sowed into a 6-well ultralow attachment plate (Corning Glass) received indicated treatment and transfection and were incubated in DMEM deprived of serum comprising 20 ng/ml epidermal growth factor (EGF), 10 ng/ml human fibroblast growth factor basic (hFGFb) as well as 2% B27. Under a light microscope (Thermo Fisher Scientific), the formed spheres were calculated 10 days later.

Measurement of aldehyde dehydrogenase (ALDH) activity

Following the standard procedure of ALDH assay kit (Sigma-Aldrich), ALDH concentration was examined.

Immunofluorescence staining

After being immobilized by 4% paraformaldehyde and perforated with 0.1% Triton X-100 for 10 min, SW620 and DLD-1 cells were treated by 5% BSA blocking buffer and then cultivated with antibodies against octamer-binding transcription factor 4 (OCT4; cat. no. ab181557; 1/250; Abcam) overnight at 4 °C, prior to the addition of Alexa-Fluor 488-conjugated secondary antibody (cat. no. #S0018; 1/100; Affinity Biosciences) for another 1 h at room temperature on the following day. The fluorescence intensity was captured under the fluorescence microscope after the nuclear DNA was dyed by 1 mg/ml DAPI for 10 min.

Reverse transcription-quantitative PCR (RT-qPCR)

With the aid of NucleoZOL® RNA Isolation Reagent (MACHEREY–NAGEL, Germany), total RNA was prepared from cells as described in the manufacturer’s manual, which was then reverse-transcripted into cDNA through SuperScript Reverse Transcriptase Kit (Vazyme, Nanjing, China). PCR reaction was implemented using LightCycler® 480 SYBR Mix (Roche Diagnostics Ltd., Shanghai, China). In terms of 2−ΔΔCq approach, DDX10, OCT4, Nanog and ATG10 expression were calibrated, with GAPDH as a normalizer.

Western blot

After isolated by RIPA buffer (BioRad, Hercules, CA, USA), quantified by BCA method (4 A Biotech, Beijing, China) and separated using gel electrophoresis, the proteins were loaded onto the PVDF membranes sealed in 5% BSA. The membranes were then processed with primary antibodies (Abcam) as well as HRP-linked secondary antibody (Abcam). By means of UltraSigna Supersensitive ECL Substrate Kit (4 A Biotech, Beijing, China), color development was conducted.

Statistics

Statistically significant differences were determined using One-way ANOVA as well as Tukey’s post hoc test, which was done with SPSS 22.0 software (SPSS, Inc., Chicago, IL, USA). P-values less than 0.05 were denoted significant in statistics.

Results

DDX10 demonstrates up-regulated expression in CRC cells

With the aim of illustrating the role of DDX10 in CRC, DDX10 expression in CRC cells was initially examined and the results from RT-qPCR and Western blotting hinted that DDX10 expression at both mRNA and protein level was conspicuously elevated in CRC cell lines (SW620, DLD-1 and LS174T) by contrast with in normal human intestinal epithelial cell line NCM460 (Fig. 1A-B). Consequently, SW620 and DLD-1 cells were selected for the follow-up experiments, in which the highest DDX10 expression was exhibited.

Fig. 1
figure 1

DDX10 demonstrates up-regulated expression in CRC cells. (A) RT-qPCR and (B) Western blotting tested DDX10 expression in CRC cells. *P < 0.05, ***P < 0.001

Full size image

Deletion of DDX10 curtails the proliferation and promotes the apoptosis of CRC cells

Subsequently, functional experiments were carried out to specify the effects of DDX10 on the malignant process of CRC. Following the transduction of siRNA-DDX10-1/2 plasmids, DDX10 expression was discovered to be tremendously depleted in SW620 and DLD-1 cells (Fig. 2A-B). Moreover, siRNA-DDX10-1 was applied to the ensuing assays in view of the lower DDX10 expression in the siRNA-DDX10-1 group. As illustrated by EDU staining, the proliferation of SW620 and DLD-1 cells was retarded overtly after DDX10 was knocked down (Fig. 2C). Conversely, through TUNEL staining, it was noted that DDX10 deficiency resulted in the significant increase on the number of apoptotic cells (Fig. 2D). The data from Western blotting also manifested that when DDX10 was down-regulated, BCL2 expression was declined and Bax expression was improved in SW620 and DLD-1 cells (Fig. 2E).

Fig. 2
figure 2

Deletion of DDX10 curtails the proliferation and promotes the apoptosis of CRC cells. (A-B) Transfection efficacy of DDX10 interference plasmids. (C) EDU staining estimated cell proliferation. (D) TUNEL staining appraised cell apoptosis. (E) Examination of BCL2 and Bax expression by Western blotting. ***P < 0.001

Full size image

Deletion of DDX10 inhibits the stemness of CRC cells

Cell stemness has been proposed as a culprit in the aggressive process of CRC. To comprehensively expound the influences of DDX10 on CRC, cell stemness was also appraised. The results from sphere formation assay delineated that after DDX10 was silenced, the sphere formation capacity of SW620 and DLD-1 cells was notably extenuated (Fig. 3A). Besides, ALDH concentration was depleted in SW620 and DLD-1 cells transfected with DDX10 interference plasmids (Fig. 3B). In addition, through RT-qPCR, Western blotting and immunofluorescence staining, OCT4 and Nanog expression were both discovered to be on a downward trend by shortage of DDX10 (Fig. 3C-E).

Fig. 3
figure 3

Deletion of DDX10 inhibits the stemness of CRC cells. (A) Sphere formation assay evaluated cell stemness. (B) ALDH assay kit tested ALDH content. Examination of OCT4 and Nanog expression by (C) RT-qPCR and (D) Western blotting. (E) Immunofluorescence staining examined OCT4 expression. ***P < 0.001

Full size image

DDX10 interference triggers ATG10-dependent autophagy

ATG genes including ATG3, ATG7 and ATG10 were predicted to be negatively regulated by DDX10 by Biogrid database. Via Western blotting, when DDX10 was down-regulated, ATG3, ATG7 and ATG10 expression were all raised, among which ATG10 was the most prominently altered (Fig. 4A). Correspondingly, ATG10 was chosen for the following assays. To verify whether the downstream mechanism of DDX10 was associated with ATG10, ATG10 expression was overtly lowered after transfection of siRNA-ATG10-1/2 (Fig. 4B-C). Furthermore, it turned out that LC3, Beclin1 expression were both increased and P62 expression was reduced by DDX10 absence in SW620 and DLD-1 cells, which were all reversed after ATG10 was depleted. Also, interference with ATG10 alone repressed LC3 expression and enhanced P62 expression in SW620 and DLD-1 cells (Fig. 4D-E).

Fig. 4
figure 4

DDX10 interference triggers ATG10-dependent autophagy. (A) Examination of ATG3, ATG7 and ATG10 expression by Western blotting. (B-C) Transfection efficacy of ATG10 interference plasmids. (D) SW620 and DLD-1 cells were co-transfected with siRNA-DDX10 and siRNA-ATG10 utilizing EZ Trans Lipo transfection reagent. Immunofluorescence staining examined LC3II expression. (E) SW620 and DLD-1 cells were co-transfected with siRNA-DDX10 and siRNA-ATG10 utilizing EZ Trans Lipo transfection reagent. Examination of LC3II/LC3I, P62 and Beclin1 expression by Western blotting. *P < 0.05, **P < 0.01, ***P < 0.001

Full size image

ATG10-dependent autophagy mediated by DDX10 inhibition hampers the proliferation and aggravates the apoptosis of CRC cells

To prove that DDX10 might function in CRC through regulating ATG10-dependent autophagy, autophagy inhibitor 3-MA was applied and ATG10 was also silenced. From the results of EDU staining, it was observed that the weakened proliferative ability of SW620 and DLD-1 cells imposed by DDX10 reduction was strengthened by ATG10 interference or 3-MA (Fig. 5A). Moreover, when DDX10 was down-regulated, the apoptotic rate of SW620 and DLD-1 cells was elevated, BCL2 expression was lowered and Bax expression was promoted. After knockdown of ATG10 or treatment with 3-MA in DDX10-silencing SW620 and DLD-1 cells, the cell apoptotic rate was repressed again, coupled with the ascending BCL2 expression and the descending Bax expression (Fig. 5B-C).

Fig. 5
figure 5

ATG10-dependent autophagy mediated by DDX10 inhibition hampers the proliferation and aggravates the apoptosis of CRC cells. SW620 and DLD-1 cells were co-transfected with siRNA-DDX10 and siRNA-ATG10 utilizing EZ Trans Lipo transfection reagent or SW620 and DLD-1 cells were transfected with siRNA-DDX10 and treated by 3-MA. (A) EDU staining estimated cell proliferation. (B) TUNEL staining appraised cell apoptosis. (C) Examination of BCL2 and Bax expression by Western blotting. **P < 0.01, ***P < 0.001

Full size image

ATG10-dependent autophagy mediated by DDX10 inhibition suppresses the stemness of CRC cells

As portrayed in Fig. 6A, sphere formation assay substantiated that the reduced size and number of spheres in DDX10-silencing SW620 and DLD-1 cells rose again by 3-MA treatment or deletion of ATG10 (Fig. 6A). 3-MA treatment or ATG10 reduction improved ALDH content again which was declined in SW620 and DLD-1 cells on account of DDX10 reduction (Fig. 6B). Similarly, the down-regulated OCT4 and Nanog expression attributed to DDX10 interruption in SW620 and DLD-1 cells were both up-regulated again after inactivation of autophagy or deficiency of ATG10 (Fig. 6C-E).

Fig. 6
figure 6

ATG10-dependent autophagy mediated by DDX10 inhibition suppresses the stemness of CRC cells. SW620 and DLD-1 cells were co-transfected with siRNA-DDX10 and siRNA-ATG10 utilizing EZ Trans Lipo transfection reagent or SW620 and DLD-1 cells were transfected with siRNA-DDX10 and treated by 3-MA. (A) Sphere formation assay evaluated cell stemness. (B) ALDH assay kit tested ALDH content. Examination of OCT4 and Nanog expression by (C) RT-qPCR and (D) Western blotting. (E) Immunofluorescence staining examined OCT4 expression. *P < 0.05, **P < 0.01, ***P < 0.001

Full size image

Discussion

CRC is commonly termed as a heterogeneous disease, the tumorigenesis and deterioration of which involve a multitude of genetic and epigenetic aberrations. In the present work, DDX10 was found to be overexpressed in CRC cells and absence with DDX10 impaired the proliferation, stemness and exacerbated the apoptosis of CRC cells, the modulatory mechanism of which might be driven by activation of autophagy mediated by ATG10.

DDX dysregulation has been reported to engage in splicing, mRNA nuclear output, and mRNA degradation to translation and alter the expression of certain oncogenes or anti-oncogenes to impact the behaviors of cancer cells (Kouyama et al. 2019; Fuller-Pace 2013). A great body of studies have claimed that DDX chiefly plays cancerous roles in CRC. For instance, DDX54 may acts as an activator of p65 and AKT signaling to facilitate the proliferation and mobility of CRC cells (Yu et al. 2021). DDX39B contributes to the metastasis of CRC via upregulating FUT3 expression and stimulating the TGFβ/SMAD2 signaling (He et al. 2021). As a member of DDX family, DDX10 expression is increased in CRC tissues and may splice RPL35 to aggravate the proliferation and metastasis of CRC cells (Zhou et al. 2022). Consistently, our research firstly presented that DDX10 expression was raised in CRC cells relative to normal human intestinal epithelial cell line and interference with DDX10 notably halted the proliferation of CRC cells. Liu et al. have held the view that inhibition of DDX10 hinders the proliferation while accelerates the apoptosis of lung adenocarcinoma cells (Liu et al. 2021). Expectedly, our experimental findings delineated that when DDX10 was depleted, the apoptotic ability of CRC cells was robustly intensified, accompanied with the elevated pro-apoptotic Bax expression and the lowered anti-apoptotic BCL2 expression. As aforementioned above, DDX10 might accelerate the occurrence of CRC.

CSCs are a subpopulation of cancer cells featured by the self-renewal and differentiation capacities similar to stem cells. Numerous studies have broadly clarified that cancer stemness represents a major source of the tumorigenesis, metastasis, recurrence and treatment resistance of CRC (Du et al. 2022; Ebrahimi et al. 2023). OCT4 and Nanog have been regarded to sustain the pluripotency of stem cells, thence being viewed as crucial regulators of stem cells (Hatina et al. 2022). ALDH is also known as a highly robust CSC marker across the vast majority of cancer types (Toledo-Guzmán et al. 2019). Here, DDX10 down-regulation was confirmed to diminish the stem-like properties of CRC cells, which was also evidenced by the declined ALDH activity, OCT4 and Nanog expression.

Interestingly, the data from Biogrid database predicated that DDX10 had a negative correlation with ATG3, ATG7 and ATG10. Our investigation further validated that ATG3, ATG7 and ATG10 expression were all enhanced in CRC cells after DDX10 was silenced. In particular, ATG10 was the most notably up-regulated. As an autophagic E2-like enzyme, ATG10 can assemble ATG12 and regulate the transformation of LC3-I to LC3-II by binding with ATG7, which is dispensable for the formation of autophagosome (Kaiser et al. 2013). P62/SQSTM1 is a multifunctional adapter protein which can interact with polyubiquitinated proteins through the ubiquitin-associated domain, further participating in the autophagic degradation of ubiquitinated substrates (Lamark et al. 2017). Beclin1 is a central protein in autophagy that triggers a cascade of proteins involved in autophagosome formation (Prerna and Dubey 2022). It is worth mentioning that DDX10 can be degraded by the activation of selective autophagy mediated by P62 (Li et al. 2023). The current experimental results uncovered that DDX10 knockdown resulted in the increase on LC3II, Beclin1 expression as well as the decrease on P62 expression in CRC cells, which were all reversed after ATG10 was depleted, highlighting that DDX10 deficiency might lead to the activation of autophagy through up-regulating ATG10 in CRC cells. Plenty studies have documented that autophagy is closely associated with the advancement of CRC and targeting autophagy may be considered as a potential therapeutic approach for CRC (Zhou et al. 2016; Zhang et al. 2023; Gil et al. 2016). Recent evidence has supported that ascending LC3B and p62 expression predicts favorable overall survival of CRC patients, whereas Beclin-1 exerts the opposite impacts (Hu et al. 2022). Specifically, ATG10 has also been determined as a biomarker for predicting the outcome of CRC patients (Hu et al. 2022) and to be implicated in the drug resistance and metastasis of CRC cells (Sun et al. 2020; Jo et al. 2017). Our study demonstrated that interference with ATG10 alone repressed LC3 expression and enhanced P62 expression in CRC cells and further testified that the impacts of DDX10 silencing on the proliferation, apoptosis as well as stemness of CRC cells were all partially abrogated following ATG10 deletion or treatment with autophagy inhibitor 3-MA.

Conclusion

All in all, this study revealed the anti-tumor role of DDX10 interference in CRC. The anti-proliferation, anti-stemness and pro-apoptosis activities of DDX10 inhibition might be associated with the activation of autophagy mediated by up-regulation of ATG10. Therefore, targeting autophagy dependent on DDX10 might be valued as a potential novel strategy to treat CRC.