This experiment was authorized by the Ethics Committee of Sichuan Provincial People’s Hospital and implemented in accordance with the Helsinki Declaration. RCC tissues were obtained from 104 RCC patients (45.0 ± 12.3 years old) who underwent surgical excision but did not receive chemotherapy or radiotherapy. The matched paracancerous normal tissues were also obtained from the same patients, which were 3–4 cm away from the tumorous tissues, and no cancer cells were discovered in the postoperative pathological examination. All tissues were frozen in liquid nitrogen and stored at 80°C for subsequent testing. All patients participating in this study signed informed consent. The general information of patients is shown in Tables 1, 2.

RCC cell lines 786-O and A-498 were bought from the America Type Culture Collection (ATCC, Manassas, VA, United States). Cells were grown in the DMEM (Thermo Fisher Scientific, Waltham, MA, United States) comprising 10% fetal bovine serum, 100 U/mL penicillin (Invitrogen, Carlsbad, CA, United States) and 100 μg/ml streptomycin (Invitrogen) and maintained at 37°C in a 5% CO₂ incubator. Cells in the logarithmic growth stage were trypsinized before being seeded into 6-well plates (5 × 10⁶ cells/well). The transfection was made after cell growth was stable. 786-O and A-498 cells were transfected with HECTD2 overexpression and knockdown plasmids, HIF-1α overexpression plasmids, miR-320a mimics/inhibitors and the corresponding negative controls. Afterward, they were incubated at 37°C with 5% CO₂. After transfection for 24 h, the cells were employed for subsequent experiments. The knockdown sequences of HECTD2: si-HECTD2, AUA​AAA​CUG​CUG​AAA​GUG​GAA; si-NC: TTC​TCC​GAA​CGT​GTC​ACG​T.

Cells were inoculated into 6-well plates (500 cells/well), and the culture solution was altered every 5 days. After incubation for 7–14 days, cells were rinsed with PBS, immobilized with 4% paraformaldehyde, and dyed with 0.5% crystal violet. At last, the colony formation number was counted using Olympus microscope (BX23, Japan).

The BrdU cell proliferation detection kit (Wuhan AmyJet Scientific Inc. Wuhan, China) was applied for cell proliferation assessment. 786-O and A-498 cells were seeded into 96-well plates (6 × 10⁴ cells/mL), respectively and incubated overnight with the constant temperature at 37°C and 5% CO₂ environment. The next day, each well was supplemented with 20 μL BrdU (50 mM) and incubated for another 2 h. The cells were fixed by 4% Paraformaldehyde for 30 min at room temperature, and incubated with peroxidase-coupled anti-BrdU-antibody (Sigma-Aldrich) for 90 min at 37°C. After being cleaned with PBS, the cells were incubated with DAPI (Beyotime, Shanghai, China) for 5 min at temperature. Subsequently, the BrdU-positive cells were counted under a fluorescence microscope (Olympus, Japan).

Cell migration and invasion were determined by the Transwell assay. Cells were put in the Transwell upper chamber at 2 × 10⁴ cells/well, and the lower chamber was supplemented with 600 μL culture medium containing 20% FBS and cultured at 37°C. Twelve hours later, the upper chamber cells were removed, immobilized with 4% paraformaldehyde, dyed with 0.1% crystal violet, dried, photographed and quantified. The invasion assay was the same as the migration assay, except that the upper chamber of the Transwell was pre-coated with Matrigel before adding the cells.

The Annexin V-FITC/PI Apoptosis Detection Kit (Yeason, Beijing, China) was used for evaluating the apoptosis of RCC cells. Briefly, the RCC cells were detached with 0.25% trypsin without EDTA and collected via centrifugation (300 g, at 4°C for 5 min). Next, the cells were washed twice with pre cooled PBS and centrifuged at 4°C for 5 min (300 g). 100 μL 1 × Binding buffer was used to suspend the cells, followed by incubating with 5 μL Annexin V-FITC and 10 μL PI stabilizing solution. After a 10-min incubation at room temperature, 400 μL 1 × Binding Buffer was added and the apoptosis was detected by flow cytometry at 488 nm (BD Biosciences, Franklin Lakes, NJ, United States).

Total RNA was separated from 786-O and A-498 cells by applying the Trizol reagent. RNA was reversely transcribed into cDNA with the PrimeScript™ RT Reagent kit (Invitrogen, Shanghai, China) per the manufacturer’s guidelines. qPCR was implemented using the Bio-Rad CFX96 quantitative PCR system and SYBR, with pre-denaturation at 95°C for 5 min, followed by denaturation at 95°C for 15 s and annealing at 60°C for 30 s. U6 served as an endogenous control of miR-320a, and GAPDH was that of HECTD2 and HIF-1α. The 2 ^(−ΔΔCt) method was adopted for statistics. Each test was done three times. Primer sequences are exhibited in Table 3.

The cells were treated, and the culture medium was removed. Then, the protein lysate (Roche) was added for total protein separation. 50 μg total protein was subjected to 12% polyacrylamide gel and went through 2 h of electrophoresis at 100 V. It was then transferred to polyvinylidene fluoride (PVDF) membranes. After being blocked with 5% skimmed milk at RT for 1 h, the membranes were rinsed with TBST three times (10 min each) and incubated with the primary antibodies (1:1,000) of anti-Bax (ab182734), anti-P53 (ab32389), anti-Caspase3 (ab32351), anti-P21 (ab109199), anti-HIF-1α (ab179483), anti-HECTD2 (ab249770), anti-E-cadherin (ab40772), anti-Vimentin (ab92547), and anti-N-cadherin (ab76011) overnight at 4°C. After washing the membranes with TBST, we incubated them with horseradish peroxidase (HRP)-labeled Goat anti-Rabbit IgG (ab205718) (1:2,500) for 1 h at RT. The above antibodies were obtained from Abcam (Cambridge, United Kingdom). Afterward, the membranes were cleaned with TBST 3 times (10 min each). Finally, the ECL chromogenic agent (Millipore, Bedford, MA, United States) was employed for exposure and imaging with a membrane scanner.

PmirGLO vectors (Promega, United States) containing miR-320a sequences with wild-type or mutant binding locations for HECTD2 (HECTD2-WT, HECTD2-MT) were adopted. 786-O and A-498 cells were co-transfected with miR-320a mimics and the luciferase vectors. Detection of luciferase activity was made with a Dual-luciferase Reporter Gene Assay System (Promega, United States).

RIP was implemented with Anti-AGO2 (#03-110, Millipore, Germany) by applying the Magna RIP RNA binding protein immunoprecipitation kit (Millipore, Germany). The RNA binding complex was analyzed by qRT-PCR. Anti-IgG served as an isotype control.

BALB/c-nu nude mice (six-week-old) were chosen to construct an in vivo tumorigenic model. Logarithmic growth phase 786-O and A-498 cells were obtained by 0.25% trypsin and resuspended in the serum-free medium to fabricate single-cell suspensions (3 × 10⁷ cells/mL). Next, 0.1 ml cell suspension was injected subcutaneously into the left forelimb armpit of each nude mouse (70 mice in total, 10 mice in each group). After injection, tumors were taken from mice under anesthesia on day 7, 14, 21, 28, and 35, respectively and weighed. The longest diameter of the tumor 1) and the shortest diameter perpendicular to it 2) were monitored with vernier calipers, and the tumor volume V (mm³) = 0.5 × a × b². All experiments were authorized by the Animal Experimentation Ethics Review Committee of the Sichuan Provincial People’s Hospital. Experiments were implemented strictly following the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (NIH Publication No. 8023).

As described previously, paraffin sections of tumor tissues were made, dewaxed with toluene, and then dehydrated with different gradients of ethanol following the standard procedure. TUNEL staining of tumor tissue was then performed using a fluorescent assay kit (Cat# 11684817910, Roche, Shanghai, China) according to the manufacturer’s guidelines.

RCC tissue specimens were sectioned by conventional paraffin embedding (4 μM), dewaxed with conventional xylene, hydrated with gradient ethanol, and inactivated with 3% H₂O₂ for 10 min. Microwave repair was made by applying 0.01 mol/L sodium citrate buffer (pH = 6.0, 15 min). Afterward, the sections were closed with 5% bovine serum albumin (BSA) for 20 min, followed by incubation with the Ki67 antibody (ab15580, Abcam, MA, United States) overnight at 4°C. On day 2, the Goat-anti-Rabbit IgG was added and incubated at RT for 20 min, cleaned with PBS and imaged with DAB. The sections were inhibited by hematoxylin, dehydrated, and transparentized, and the closure was observed under the microscope.

Paraffin tissue sections were dewaxed and hydrated. After antigen repair, the sections were sealed and maintained with the primary antibodies, including anti-HIF-1α (ab179483, Abcam) and anti-HECTD2 (HPA037767, Sigma-Aldrich). After being washed with PBS, the Goat polyclonal Secondary Antibody to Rabbit IgG-H&L (Alexa Fluor® 488) (ab150077) was added for incubation. Afterward, 4,6 diamidinyl-2-phenyl indole (DAPI) nuclear staining was conducted, and the sections were blocked, observed, and photographed. Three sections were chosen from each specimen, and five non-overlap high-power fields of cancer tissues were randomly chosen from each section. The number of HIF-1α and HECTD2 staining positive cells in each field of view was counted using Image-Pro Plus image analysis software, and the number of HIF-1α and HECTD2 staining positive cells/mm² was calculated.

SPSS22.0 statistical software (SPSS Inc., Chicago, IL, United States) was employed for statistical analysis. All data were presented as mean ± SD. Measurement data were expressed as mean ± standard deviation (x ± s), and enumeration data were expressed as four tables (or percentages). χ² was adopted to compare differences between the two groups. One-way ANOVA was employed for the multi-factor comparison, and the comparison between two groups was made by t test. p < 0.05 represented statistical significance.

To investigate whether HIF-1α and HECTD2 contribute to RCC progression, we consulted the database (https://www.proteinatlas.org/) and discovered that HIF-1α and HECTD2 levels were both highly expressed in RCC tissues (Figures 1A,B). Also, WB results uncovered that HIF-1α and HECTD2 expression levels were significantly higher in tumor tissues of RCC patients than those in corresponding surrounding non-tumor tissues (Figures 1C,D). More importantly, Fuhrman grade in patients with higher profiles of HIF-1α and HECTD2 was generally staged III + IV and tumor stage tended to be pT2/T3 (Tables 1, 2). Besides, there is a positive relationship between HIF-1α expression and HECTD2 expression in KIRC (analyzed by GEPIA) (Figure 1E). Meanwhile, patients with high expression of HIF-1α and HECTD2 had worse prognosis compared with those with low expression of HIF-1α and HECTD2. Additionally, those patients with higher HECTD2+higher HIF-1α had poorer overall survival of those with higher HECTD2+lower HIF-1α (p = 0.032) [data analyzed via Kaplan-Meier Plotter (http://kmplot.com/analysis/)] (Figures 1F,G). These results prompted that HIF-1α and HECTD2 were associated with the malignant phenotype of RCC cells and might exert carcinogenic effects.

We constructed the HECTD2 overexpression model in 786-O and A-498 cell lines and tested the transfection effect by RT-qPCR to probe the influence of HECTD2 on the malignant progression of RCC (p < 0.05, Figure 2A). Meanwhile, the detection of RCC cell proliferation and apoptosis was made using the colony formation experiment, BrdU assay and flow cytometry, respectively. Interestingly, overexpressing HECTD2 significantly motivated proliferation and bridled apoptosis in RCC (p < 0.05, Figures 2B–D). Additionally, WB results illustrated that the pro-apoptotic proteins P53, Bax, p21, and c-Caspase3 were significantly down-regulated after HECTD2 overexpression (p < 0.05, Figure 2E). Besides, Transwell assay results testified that overexpressing HECTD2 significantly facilitated RCC cell migration and invasion (p < 0.05, Figure 2F). Also, WB results signified that overexpressing HECTD2 significantly repressed the expression of the epithelial marker E-cadherin and elevated the levels of interstitial markers Vimentin and N-cadherin (p < 0.05, Figure 2G). These findings confirmed that overexpressing HECTD2 intensified the malignant progression of RCC.

We constructed a HECTD2 overexpression model in 786-O and A-498 cells to testify the influence of HECTD2 on tumor growth in vivo and found that overexpressing HECTD2 significantly heightened tumor volume and weight (p < 0.05, Figures 3A–C). Also, IHC revealed a substantial elevation in the Ki67-positive cell number in 786-O and A-498 cells after HECTD2 overexpression (p < 0.05, Figure 3D). Moreover, WB results demonstrated that overexpressing HECTD2 in 786-O and A-498 cells significantly facilitated the HECTD2 profile in the transplanted tumor tissues (p < 0.05, Figure 3E). The tissue immunofluorescence data exhibited that the fluorescence intensity of the HECTD2 group was more obvious than that of the control group (p < 0.05, Figure 3F). These results added to evidence that overexpressing HECTD2 boosted RCC cell growth.

To ascertain the function of HIF-1α and HECTD2 in RCC evolvement, we knocked down HECTD2 on the basis of HIF-1α overexpression in 786-O cells and monitored the expression of HIF-1α and HECTD2 using RT-qPCR. It turned out that HIF-1α and HECTD2 were up-regulated after HIF-1α overexpression. However, compared with the HIF-1α group, the HIF-1α expression in the HECTD2 group was not changed, while HECTD2 was evidently down-regulated (p < 0.05, Figure 4A). The colony formation, BrdU assay and flow cytometry were employed to examine RCC cell proliferation and apoptosis. The results unveiled that overexpressing HIF-1α facilitated the proliferation and impeded the apoptosis of RCC. Nevertheless, the findings of the HIF-1α + si-HECTD2 group were exactly opposite to those of the HIF-1α group (p < 0.05, Figures 4B–D). Meanwhile, Transwell assay results clarified that HIF-1α overexpression dramatically strengthened RCC cell invasion and migration. In contrast, RCC cell migration and invasion were significantly hampered in the HIF-1α + si-HECTD2 group (vs. the HIF-1α group) (p < 0.05, Figure 4E). WB exhibited that the expression of pro-apoptotic proteins P53, Bax, p21 and c-Caspase3 were suppressed after overexpressing HIF-1α. However, the profiles of these proteins were heightened in the HIF-1α + si-HECTD2 group (vs. the HIF-1α group) (p < 0.05, Figure 4F). Also, WB results confirmed that HIF-1α overexpression significantly hampered the expression of E-cadherin and elevated the levels of Vimentin and N-cadherin (vs. the control group). The addition of si-HECTD2 to HIF-1α subverted the effect of HIF-1α (p < 0.05, Figure 4G). These results manifested that up-regulating HIF-1α in 786-O cells resulted in the elevation of HECTD2 profiles and the aggravation of RCC.

To further validate the impact of the HIF-1α/HECTD2 axis on tumor growth in vivo, we established HIF-1α overexpression and/or HECTD2 knockdown models in 786-O cells and discovered that overexpressing HIF-1α increased tumor volume and weight (vs. the vector group). However, HECTD2 knockdown suppressed RCC tumor volume and weight (vs. the HIF-1α group) (p < 0.05, Figures 5A–C). Also, immunohistochemical results exhibited evident facilitation in the Ki67-positive cell number in 786-O cells after HIF-1α overexpression, whereas the effect was reversed after HECTD2 knockdown (p < 0.05, Figure 5D). Besides, WB results illustrated that the profiles of HIF-1α and HECTD2 were elevated after HIF-1α overexpression in 786-O cells. However, compared with the HIF-1α group, the HIF-1α expression remained unchanged in the HIF-1α + si-HECTD2 group, but HECTD2 was significantly down-regulated (p < 0.05, Figure 5E). The results of tissue immunofluorescence exhibited that the fluorescence intensity of HIF-1α and HECTD2 was more pronounced in the HIF-1α group (vs. the vector group). Nevertheless, compared with the HIF-1α group, the fluorescence intensity of HIF-1α in the HIF-1α + si-HECTD2 group showed little change, but the fluorescence intensity of HECTD2 was dramatically weakened (p < 0.05, Figures 5F,G). These findings further confirmed the influence of HIF-1α overexpression on HECTD2 and RCC cell growth in vivo.

For analyzing the upstream mediator of HECTD2, we analyzed the positive or negative related genes of HECTD2 via LinkedOmics, and those genes were subjected to GSEA enrichment analysis. The results indicated that HECTD2 is involved in a KEGG pathway of MicroRNAs in Cancer (Figure 6A). Inspired by the miRNA-mRNA regulatory network, we adopted Targetscan, miRmap, microT, PicTar and miRanda to search the miRNA target of HECTD2 and discovered that HETCD2 had seven miRNAs as its common targets (Figure 7A). After that, we examined the profiles of these seven miRNAs in HIF-1α overexpressing cells by RT-PCR. It turned out that miR-320a was most significantly down-regulated (Figure 6C). The base binding sites between miR-320a and HECTD2 was shown in Figure 6D. The dual-luciferase reporter assay results demonstrated that miR-320a weakened the luciferase activity of HECTD2-WT, but it had no substantial effect on HECTD2-MUT (p < 0.05, Figure 6E). The RIP results demonstrated that miR-320a mimic transfection resulted in a higher amount of HECTD2 deposited in the Ago2 antibody group than that in the IgG group, implying that HECTD2 was bound to Ago2 via miR-320a (p < 0.05, Figure 6F). Moreover, the effect of modulating miR-320a on HECTD2 expression was examined by RT-qPCR. Interestingly, overexpressing miR-320a hampered the HECTD2 profile, while miR-320a knockdown showed the reverse (p < 0.05, Figure 6G). These findings testified that there was a targeting correlation between miR-320a and HECTD2.

To figure out the influences of miR-320a on HECTD2 and RCC evolvement, we transfected miR-320a mimics in A-498 cells and verified the transfection validity by RT-qPCR (p < 0.05, Figure 7A). Meanwhile, the results of the colony formation assay, BrdU assay and flow cytometry illustrated that overexpressing miR-320a restrained RCC cell proliferation and intensified apoptosis (p < 0.05, Figures 7B–D). Additionally, Transwell assay results manifested that RCC cell migration and invasion were curbed by miR-320a overexpression (p < 0.05, Figure 7E). Besides, WB results uncovered that the transfection of miR-320a mimics up-regulated P53, Bax, p21, and c-Caspase3 (vs. miR-NC group) (p < 0.05, Figures 7F,G). Moreover, WB results also signified that overexpressing miR-320a significantly facilitated E-cadherin expression and hampered the levels of Vimentin and N-cadherin (vs. the miR-NC group) (p < 0.05, Figure 7H). These results testified that transfection of miR-320a mimics in A-498 cells alleviated RCC progression.

To explore the specific mechanism by which miR-320a affected RCC growth and EMT, we transfected miR-320a mimics in 786-O cells overexpressing HECTD2 and assayed the transfection validity by applying RT-qPCR. As a result, compared with the NC group, miR-320a was down-regulated in the HECTD2 group and was up-regulated in the miR-320a group (p < 0.05, Figure 8A). The colony formation assay, BrdU assay and flow cytometry results exhibited enhanced RCC cell proliferation and significantly repressed apoptosis in the HECTD2 group compared to the NC group, while the transfection of miR-320a mimics exerted the opposite function (p < 0.05, Figures 8B–D). Meanwhile, Transwell assay results showed that RCC migration and invasion were significantly enhanced in the HECTD2 group compared with the NC group, while they were attenuated after the miR-320a mimic transfection (p < 0.05, Figure 8E). Also, WB exhibited that the protein profiles of P53, Bax, p21 and c-Caspase3 were evidently declined in the HECTD2 group (vs. the NC group). Nevertheless, these proteins were significantly up-regulated in the HECTD2 +miR-320a group (vs. the HECTD2 group) (p < 0.05, Figure 8F). Moreover, WB results illustrated that the E-cadherin profile was impeded, while Vimentin and N-cadherin were significantly up-regulated in the HECTD2 group (vs. the NC group). The HECTD2+miR-320a group had completely opposite results to the HECTD2 group (p < 0.05, Figure 8G). These findings manifested that HECTD2 enhanced RCC cell growth and EMT in 786-O cells, but transfection with miR-320a mimics significantly alleviated RCC (Graphical Abstract).